ML20137M681
| ML20137M681 | |
| Person / Time | |
|---|---|
| Issue date: | 01/13/1986 |
| From: | Stello V NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | |
| References | |
| SECY-86-010, SECY-86-10, NUDOCS 8601280338 | |
| Download: ML20137M681 (157) | |
Text
_ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _
, 'r tlA LECQ #
,/ %,
l 3- :E I
\...../
POLICY ISSUE January 13, 1986 SECY-86-10 For: The Commissioners From: Victor Stello, Jr.
Acting Executive Director for Operations
Subject:
RECOMMENDATIONS FOR IMPROVING TECHNICAL SPECIFICATIONS
Purpose:
To inform the Commissioners of the recommendations of the Technical Specification Improvement Project (TSIP) and planned staff actions in response to these recommendations.
Background:
In March 1982, the NRC proposed a change to 10 CFR 50.36 to focus Technical Specifications on items of greatest safety importance. The proposed approach was to split existing requirements between two documents, the plant Technical Specifications which would remain an appendix ~to the operating license and a separate set of supplemental requirements of lesser safety significance which would not be a part of the operating license. Because of difficulties with defining the criteria for making the split and other higher priority licensing work, implementation of the rule change was delayed.
In August 1983, the Office of the Executive Director for Operations established a Task Group on Technical Specifications to identify the scope and nature of problems with surveillance testing in current Technical Specifications and to develop alternative approaches that will provide better assurance that surveillance testing does not adversely impact safety. The product of this Task Group was NUREG 1024 " Technical Specifications -
Enhancing the Safety Impact."
On December 31, 1984, NRR established a Technical Specification Improvement Project (TSIP) to reconsider the entire subject of Technical Specifications and provide recommendations for improvement. TSIP work was coordinated CONTACT: Edward J. Butcher x24559 8601280338 860113 PDR SECY 86e010 PDR
, e closely with a simi.lar industry effort sponsored by the Atomic Industrial Forum (AIF). The TSIP and AIF final reports are provided as enclosures.
Discussion: The principal finding of the TSIP Report is that tnare are no acute safety concerns or resource burdens associated with Technical Specifications which woula support imposing a mandatory program of changes to the Technical Specifications of operating reactors. There were, however, important problem areas identified where significant improvements could best be realized through a cooperative program between the industry and NRC to revise the existing Standard Technical Specifications (STS). The problem areas can be summarized as follows:
- 1. Lack of well-defined criteria for what should be included in Technical Specifications, and
- 2. Human factors and other technical weaknesses in the Technical Specifications.
The first problem, in addition to a reluctance of the NRC staff to use tools other than the Technical Specifications for implementing regulatory requirements.,
has resulted in the Technical Specifications becoming a catch-all for staff requir'ements and,-thus, covering too wide a range of safety importance. In the second problem area specific problems of clarity, poor wording and inadequate statement of the bases were also identified in NUREG-1024. The result has been a document which is voluminous and difficult to use.
To address these problems TSIP recommended that:
- 1. A Commission Policy Statement be issued which defines the scope and purpose of Technical Specifications and encourages licensees to implement a program to upgrade their Technical
- Specifications.
- 2. The NRC staff give increased attention to changes made-by licensees using the 10 CFR 50.59 process.*
^This would be a first step in reducing the reluctance of individual ::RC staff units to use the FSAR as a tool other than the Technical Specifications for implementing regulatory requirements.
a e
- 3. The NRC staff review and revise the STS te correct human factors and other technical weaknesses through a program of outside technical assistance and dedicated in-house resources.
- 4. The NRC encourage the continued development of probabilistic risk assessment (PRA) methods to address Technical Specification requirements.
Specifi.cally, with regard to the proposed Policy Statement, TSIP recommended that the Commission adopt as policy for all power reactors the fundamental principles of importance and immediacy embodied in the Atomic Safety and Licensing Appeal Board decision ALAB-531 which states:
" Technical Specifications are to be reserved for those matters as to which imposition of rigid conditions or limitations upon reactor operation is deemed necessary to obviate the possibility of an abnormal situation or event giving rise to an immediate threat to the public health and safety." (9 NRC-263, 1979)
A' specific set of objective criteria based on this concept was developed in conjunction with AIF for determining which systems, structures and process variables must be controlled by Limiting Conditions for Operation (LCO) in the Technical Specifications.
These criteria are:
A. An installed system that is used to detect, by monitors in the control room, a significant abnormal degradation of the reactor coolant pressure boundary, B. A process variable that is an initial condition of a DBA analysis, or C. A structure, system, or component that is part of the primary success path of a safety sequence analysis and functions or actuates to mitigate a Design Basis Accident.
Regulatory requirements which do not meet this test would be reserved for other controlled documents such as the updated FSAR or Quality Assurance Plan.
The Office of iNuclear Reactor Regulation has initiated
! actions to develop a complete program to verify the practicality of implementing the recommendations in the
, P 4
TSIP report. Validation.of the criteria as a mechanism for implementing the principles of importance and immediacy in ALAB-531 is in progress by a trial split of an existing set of PWR and BWR Technical Specifications. The balance of the program will be focused on the short term implementation of improvements to the existing STS and the longer term development of a new more streamlined set of STS based on the criteria. Some of the improvements proposed would require minor rule changes, such as deleting the requirement for Radiological Effluent Technical Specifications (RETS) (10 CFR 50.36a). However, the staff believes that accomplishment of the overall program objectives can be initiated without the major rule change proposed in 1982. Instead, a Commission Policy Statement would be issued, as recommended by TSIP, to adopt the principles of ALAB-531 and establish the specific criteria to be used for defining the scope and purpose of Technical Specifications. After experience is~ gained in the application of the criteria and implementation of the program, a decision on codification of the criteria can be made.
The staff has initiated a series of meetings with the various groups representing the industry to further identify and prioritize the short term improvements that can be made and to define the level of industry participation in both the short term and long term aspects of the program. The staff believes-that the active participation of the industry and other interested groups is a critical element in the timely achievement of our '
goal of improving Technical Specifications. To begin this process the staff met with the Chairman of the AIF Subcommittee on Technical Specification Improvements, on December 4, 1985 and with the full AIF Subcommittee on December 11,.1985. We have also begun a series of meetings with individual Owners Groups by meeting with the B&W Owners Group on December 10, 1985. By the end of January 1986, we plan to have completed at least an initial meeting with all of the Owners Groups.
At the conclusion of this first round of meetings with the various industry groups, the staff will develop a complete Program Plan for seeking appropriate approvals l and implementing specific Technical Specification t
~ improvements. By meeting with the industry before i
developing our final plan, we hope to be able to develop l anintegratedapproachwhichwillassurethemost l effective use of both industry and NRC resources in achieving the desired improvements. Our Program Plan
- P
' will include.both resource estimates and schedule goals for specific activities including the issuance of a Commission Policy Statement on Technical Specifications.
The plan will consist of activities to implement both short term improvements to the existing STS and a longer term complete rewrite of the STS based on the TSIP criteria for establishing Technical Specification requirements.
Our current schedule goal for completing the Program Plan is March 1, 1986. Implementation of the Program Plan will follow immediately. One of the initial tasks 1
will be the preparation of a Commission Policy Statement.
Other high priority areas will include implementation of the short term improvements identified in meetings with the Owners' Groups and an upgrade of the Bases sections in the existing STS. As stated above, the ultimate long '
term objective of the Program Plan will be to completly r u.'ite and streamline the STS. <
- ,f'. ' l4l .
Vi'ctor Stello, Jr.
! Acting Executive' Director for Operations
Enclosures:
- 1. TSIP Final Report
- 2. AIF Final Report DISTRIBUTION:
Commissioners OGC OPE OI OCA ACRS EDO SECY l
~
e P BACKGROUND AND SIGNATURE TAB Use this side of the sheet to precede the signature page when assembling correspondence.
(USE REVERSE SIDE FOR BACKGROUND TAB) a e
f e
j t i ,
g h 9 4 e o a
w I e
k g O e, I
.r m, m
2 i L -
p .
a C
- D m
e--, w v. .,e. + ,.- j. g ,
e eG
. c 1
RECOMMENDATIONS FOR IMPROVING
'i TECHNICAL SPECIFICATIONS SEPTEMBER 30, 1985 Technical Specification Improvement Project Don H. Beckham, Director Samuel E. Bryan, Assistant Director Timothy E. Collins Richard L. Emch, Jr'.
Scott F. Newberry Peter M. Williams Legal advice provided by Richard K. Hoefling
> a_- --- - + ---+r i
i The Technical Specification Improvement Project was established by the.
Director, Office of Nuclear Reactor Regulation on December 31, 1984. This
! report pra,ents the conclusions and recommendations of the Project to the Director, NRR and summarizes the Project efforts. The conclusions and recommendations do not represent NRR or Commission Policy.
l 1
1
'1 l
ii TABLE OF CONTENTS i
fage 1 Introduction and Summary............................................. 1 2 Conclusions and Recommendations...................................... 7 2.1 Conclusions.................................................. 7 j
, 2.2 Recommendations.............................................. 8 i
i 2.2.1 Policy Statement on Technical Specification Criteria......... 8 2.2.2 Increased Attention to 10 CFR 50.59......................... 21 2.2.3 Revisions to Technical Specification Bases.and Content...... 23 2.2.3.1 Short-Term Improvements..................................... 23 2.2.3.2 Long-Term Improvements.......................................'27 2.2.4 Use of Probabilistic Risk Assessment to Improve Technical Specifications.................................. 29-3 Impacts of Implementing Recommendations............................. 31 .
3.1 Safety...................................................... 31 3.2 Resources................................................... 32 3.2.1 Near Term................................................... 32 3.2.2 Long Term................................................... 33 3.3 Adm i n i s t r a ti o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
l l
iii i
.P a_a_e.
4 Problem Identification.............................................. 37 4.1 Use of Alternate Tools...................................... 40 4.1.1 Rol e o f R e g ul a to ry Tool s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.2 Lack of Criteria............................................ 43 4.3 Human Factors and Other Technical Weaknesses................ 45 4.3.1 Bases....................................................... 45 4.3.1.1 The Bases Problems and Their Significance. ................. 48 4.3.2 C l a r i ty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .' 5 0 5 Development and Evaluation of Alternate Solutions...................
54 5.1 Safety Impact............................................... 5'5 5.2 Resource Impact............................................. 57
. 5.3 Administrative Impact....................................... 60 5.4 Evaluation Scoring.......................................... 61 6 References...........................................................
Appendix A Background on Purpose'of Technical Specifications.........
i e
iv LIST OF TABLES AND FIGUR$5 Page i Figure 2.1 Changes to Facilities, Procedures, and Tests (or experiments).......................................... 13 I
Table 2.1 Technical Specification Remaining Under Recommended Criteria...................................... 16 Table 2.2 Specification Which Require Reactor Shutdown or Power Limitations...................................... 20
, Table 3.1 Substantive Regulations in Title 10 Involving Technical Specifications for Power Reactors............... 36 i
Table 4.1 Plant Operations Staff Interviews by TSIP Utility / Vendor Office Visits.............................. 39 E
Table 5.1 Evaluation Matrix'for Proposed Solutions to Technical Specification Problems....................... 63-70
a 1 INTRODUCTION AND
SUMMARY
In November 1983, the Task Group on Technical Specifications published NUREG 1024, " Technical Specifications - Enhancing the Safety Impact" (Sniezek, November 1983). The NUREG contained specific recommendations for improving Technical Specifications. The Director, Office of Nuclear Reactor Regulation (NRR), was directed to implement the recommendations (Dircks, November 1983). Subsequent events both delayed implementation and highlighted the need for improvements.
In December 1984, the Director, NRR, chartered the Technical Specification Improvement Project (TSIP) to reconsider the entire area of Technical specifications and provide him with recommendations and changes needed to I implement the recommendations (Denton, December 1984). TSIP was also to facilitate short-term improvements to specific areas of Technical Specificat' ions where possible.
Therefore, the project members have not only studied the problems with Technical Specifications from a detached vantage point, but have also had to deal day-to-day with the technical and institutional barriers that must be dealt with to effect improvements. The recommendations in this report were derived from both activities.
In the TSIP Program Plan TSIP stated, "The history of Technical Specifications indicates that the concerns expressed [in the plan introduction] are not new.
The efforts to date have largely been to strike a balance between identifying a set of specifications that ' enable [the Commission] to find that the utilization or production of-special nuclear material will be in accord with the common defense and security, and will provide adequate protection to the
health and safety of the public;' (Atomic Energy Act) and also wi.ll not be so detailed as to unnecessarily hamper facility operation." (Beckham, March 1985).
TSIP maintains that opinion and views its work as an effort to improve'on the past. TSIP recognized that it is not writing on a clean slate.
Starting from the discussion in NUREG 1024, TSIP collected information to identify the root causes of problems with the Technical Specifications. Its meetings with vendor and utility personnel, the Atomic Industrial Forum's (AIF)
Subcommittee on Technical Specification Improvements, and representatives of public interest groups are described in Section 4 of this report. From these meetings, its technical assistance work, and its efforts on short-term projects, T5IP developed an overwhelming sense that, although NURFG-1024 captured specific problems with Technical Specifications, it did not identify root causes. Despite the enthusiasm and support shown by utilities for correcting some of the problems identified in NUREG-1024, little progress has been ma'de since 1983. Therefore, TSIP's efforts concentrated on going beyond the compelling specific problems to the underlying foundation that was inhibiting correction of the problems.
From this work, as described in detail in Section 4, TSIP identified three majorproblems:
(1) Reluctance of the NRC staff to use tools other than Technical Specifications for establishing regulatory requirements, (2) Lack of well-defined criteria for what requirements should be included in Technical Specifications, and l ~_ _- - _ , _
-3 (3) Human factors and other technical weaknesses of the Technical Specifications.
TSIP concluded that these problems do not pose an acute safety problem for operating power reactors. .However, if the NRC is to achieve effective and efficient regulation of the nuclear industry, each problem must be addressed.
After identifying the problems, TSIP went through an extensive program of j evaluating alternatives, as described in Section 5 of the report. TSIP evaluated each alternative against two conditions. First, did it resolve all or part of the problems? Second, what were the safety impacts, the resource impacts, and the aaministrative burdens associated with the alternative? The assumptions that TSIP used are also detailed in Section 5. ,
The major difficulty facing TSIP as evidenced from the problem statements above, is a lack of consistent philosophy on the roles of the different regulatory tools in relation to safety issues of differing importance.
Therefore, although TSIP's first recommendation is that a Policy Statement
~
be prepared to articulate the scope and purpose of Technical Specifications (Recommendati.on 1), this effort could-be enhanced by improving NRC staff's understanding of the relationship of safety to the roles of regulations, orders, license conditions, Technical Specifications, and commitments made in the Final Safety Analysis Report-(FSAR).
The Commission Policy Statement should include criteria to identify Technical Specifications. In evaluating potential criteria, TSIP found itself returning I
to the subjective. criterion used by the Atomic Safety and Licensing Appeal Board in ALAB-531, "that Technical Specifications are to be reserved for
~
those matters as to which imposition of rigid conditions or limitations upon reactor operation is deemed necessary to obviate the possibility of an abnormal situation or event giving rise to an immediate threat to the public health and safety." (9 NRC 263, 1979)
To remove the subjective judgements required by that statement, TSIP attempted to identify objective criteria that would capture that subjective concept.
In pursuing these criteria, TSIP had several discussions with the AIF Subcommittee, who were pursuing a similar goal.
Out of these discussions grew a set of proposed criteria for screening items for inclusion in Technical Specifications. The proposed criteria are strongly linked to the Design Basis Accident (DBA)* analysis included in the FSAR a'nd are discussed in detail in Section 2.2.1.
Admittedly the DBA concept does not include several items imposed on licensees by regulation or order subsequent to licensing. However, TSIP feels strongly that the proposed criteria capture C.ast, if not all, of the items that would be included in the subjective criteria of " immediacy" and "most important items" that we each carry around in our heads.
" DESIGN BASIS ACCIDENT - A postulated event, analyzed in the Final Safety Analysis Report, for which a structure, system, or component must meet its functional goals. These analyses are contained in Chapters 6, 15 of the FSAR (or equivalent chapters) and are idcatified as Condition II, III, or IV Events (or equivalent) that either assume the. failure of, or challenge to the integrity, of a fission produc.t barrier.
t n.- - -,-w qg , - - -
5-Because individual items that fall inside or outside of these criteria might have the same immediacy and importance articulated by the ALAB, TSIP has recommended a Policy Statement instead of a rule. However, it should be clear that the burden of proof would fall to the group proposing to add items not consistent with the criteria, or to delete or items consistent with the criteria. And the burden should be heavy.
1 With a strict program for controlling Technical Specifications, the staff must become more familiar and comfortable with other regulatory tools. The evaluation and review process addressed by 10 CFR 50.59 is the primary mechanism recommended by TSIP for controlling items not in Technical Specifications (Recommendation 2). Currently, there is a mistrust of that mechanism because of perceived abuse by licensees. This mistrust can be .
overcome by training and by joint NRR, IE, Regional programs to audit the licensees' 50.59 processes. The mechanisms already exist in the IE Inspec' tion Manual and regional instructions. Specific recommendations for increasing NRR involvement are included in Section 2.2.2.
Finally, TSIP recognizes that the improvements recommended by NUREG-1024 and l confirmed in its work will only be made through a program involving the utilities and owners groups in the technical ~ evaluations.
(Conclusion 4, Recommendations 1, 3, and 4). TSIP believes that allowing i licensees to revise their existing Technical Specifications to reflect the proposed criteria will serve as incentive for them to improve the Bases, clarity and technical adequacy of the items that remain Technical Specifications. Because TSIP could find no acute safety concern with existing l
i
Technical' Specifications, the program should be voluntary. Therefore, industry sup'p ort is essential in realizing the improvements that can and are being made. NRC reluctance to relinquish any items from Technical Specificati6ns to other control mechanisms will surely limit the cooperation extended by licensees.
This summary is presented to give the reader an overview of the goals of the Technical Specification Improvement Project and a framework of how the recommendations support thse goals. Ultimately, the result of the TSIP would be submittals by each participating licensee of a detailed request for a change to Technical Specifications. This change request will require review, notification, and approval prior to implementation just as for any license change. The purpose of TSIP is to remove institutional barriers to I legitimate changes, to minimize the subjectivity of what constitutes a i i legitimate change, and to provide a framework for handling items not incorporated in Technical Specifications. TSIP has no vested interest in any piece of the-recommendations. For example, if criteria could be proposed that are equally objective, capture the concepts of immediacy and importance, and can be implemented without significant re-analysis of operating reactors, then TSIP would gladly support them.
But each recommendation is a piece of the overall program'that tries to balance NRC needs, industry concerns, and most importantly, improving the regulatory process to protect the public health and safety.
k 2 CONCLUSIONS AND RECOMMENDATIONS 2.1
Conclusions:
(1) There are no acute safety concerns or resource burdens which would support imposing a mandatory program of changes to the Technical Specifications of operating reactors.
]
(2) Improvements in both safety and resource requirements can be realized through a focused effort to correct human factors and other technical weaknesses in the Technical Specifications.
(3) Definition of the scope and purpose of Technical Specifications would provide useful guidance to the NRC and should lead to substantial NRC resource savings.
i t
, (4) Maximun safety enhancement will require participation of licensees and other interested grcups.
l l (5) Definition of the scope and purpose of Technical Specifications is an important incentive for industry participation in a program to improve Technical Specifications.
1
l 2.2~ Recommendations:
(1) A Commissian Policy Statement should be issued which defines the scope and purpose of Technical Specifications as discussed in Section 2.2.1 and encourages licensees to implement a program to upgrade their Technical Specifications.
(2) The NRC should give increased attention to changes made by licensees using the 10 CFR 50.59 process.
(3) The NRC should review and revise the Standard Technical Specifications to correct human factors and other technical weaknesses through a program of technical assistance and dedicated in-house. technical resources.
(4) The NRC should encourage the continued development and application of' probabilistic risk assessment methods to address Technical Specifications requirements.
2.2.1 Policy Statement on Technical Specification Cciteria A Commission Policy Statement should be issued which defines the scope and purpose of Technical Specifications and encourages licensees to implement a program to upgrade their Technical Specifications.
An essential first step in correcting the root causes of the problems identified with Technical Specifications is es'tablishing clearly the 1
1
-g-purpose and scope of Technical Specifications. Without this step, it is impossible to evaluate the adequacy of the control mechanisms used for other areas. Therefore, TSIP evaluated several previously articulated statements of. purpose. The criterion expressed by the Atomic Safety and Licensing Board in ALAB-531 was considered to be the most concise subjective statement of the purpose. That statement,
" Technical Specifications are to be reserved for those matters as to which imposition of rigid conditions or limitations upon reactor operation is deemed necessary to obviate the possibility of an abnormal situation or event giving rise to an immediate threat to the public health and safety." (9 NRC 273, 1979) should be adopted as the Commission's purpose.for Technical Specifications.
To make-that statement of purpose useful in the regulatory arena, the subjective statement of purpose has to be expressea, as clearly as possibl.e, in objective criteria that define the scope of Technical Specifications.
Although there will probably always be some gray area and discussions of some specific issues will persist no matter what criteria are developed, the proposed objective criteria should minimize that gray area.
In efforts to develop an acceptable set of objective criteria, TSIP hi.s had several interactions with the AIF Subcommittee and its Working Group on Technical Specification Criteria. Through these actions, and significant~
work by the Working Group, a set of criteria were developed that TSIP concludes captures an adequate statement of the scope and purpose of Technical Specifications. These criteria were written to be applied to the. Limiting Safety System Settings (LSSS), the Limiting Conditions for Operation (LCO),
and the Surveillance Requirements that support the LCOs. The areas of Bases, Design Features, Safety Limits and Administrative Controls are adequately defined in 10 CFR 50.36. Some inappropriate material has been included in these latter areas and this is specifically discussed later.in this Section.
The significant problems arise from the selection.of structures, systems, and
~
components included in Technical Specifications, and therefore, the criteria focus on those areas. The recommended criteria for evaluating LSSSs and-LCOs are as follows:
(1) An installed system that is used to detect, by monitors in the control room, a significant abnormal degradation of the reactor coolant pressure boundary, (2) A process variable that is an initial condition of a DBA analysis, (3) A structure, system, or component that is part of the primary success path of a safety sequence analysis and functions or actuates to mitigate a Design Basis Accident.
The AIF Subcommittee has prepared a full recommended program for implementing the Technical Specification revisions. Although TSIP has reviewed a draft of this proposal, it is not incorporating the entire program into these recommendations.
TSIP concludes that the current 10 CFR 50.36 could serve as a vehicle to solve the problems identified. The full AIF proposal would require changing the rule prior to implementation. TSIP concludes that the improvements
recommended below can proceed without a change to 10 CFR 50.36. The TSIP has reviewed each of the categories in the present rule and recommends the following.
Safety Limits and Limiting Safety System Setting 3 would be retained in their present form and content.
Limiting Conditions for Operation (LCOs) would be examined using the recommended criteria to determine which LCOs would remain in Technical Specifications. For items removed from Technical Specifications, licensees would be required to address the location and controls for the technical content of the Technical Specifications removed. Modifications to the licensee's FSAR would generally be required.
Surveillance Requirements would remain for LCOs which remain in the Technical Specifications. However, TSIP concludes that extracting Surveillance Requirements into a separate document could improve their useability and would be an acceptable alternative. The document would be incorporated into
- the Technical Specifications by reference and changes would continue to require prior NRC approval.
Design Features would be retained in its current form and content. While TSIP has doubts-about the need for all the information currently in this category, modification of the category would require a rule change and TSIP l would not recommend a rule change for.this purpose alone. This category does l
l serve in its present form to demonstrate compliance with Section 182 of the l
Atomic Energy Act.
l _ _ _ - . _ - - _
Administrative Controls would also be retained. However, TSIP concludes that current Technical Specifications contain substantial information extraneous to the purpose of this category. The Administrative Controls category should be limited to: (1) a requirement that certain procedures described in the FSAR be adherred to, (2) a description of the process utilized by the licensee to control and review changes to organization and procedures (this would
~
include thc licensee's process to assure compliance with 10 CFR 50.59), (3) the licensee's controls to assure adequate record keeping, and review and auditing of licensee activities, and (4) specific requirements, if any, for reporting to the NRC not already defined in regulations. Figure 2.1 describes the change process in place at most nuclear facilities.
All other material currently in this category of Technical Specifications would be removed. Specifically, TSIP views the sections of current Technical 4
Specifications describing procedures and dealing with Responsibility, Organization, Qualifications and Training as inappropriate for Technical Specifications. This material would be placed in the FSAR either directly or by reference and would be subject to the control of 10 CFR 50.59.
In implementing this program, TSIP recommends that the current 10 CFR 50.36 be retained. Some regulatory changes would nontheless be required but a major i
- . FIGURE 2.1 gy'lGES TO FACILITIES, PROCEDURES A!4D TESTS (OR EXPERIMENTS)
Change Proposal .
if -
15 the Safety Analysis Report (SAR) affected?
(1) Does the proposal change the facility or procedures from their description in the SAR?
(2) Does the proposal involve a test or experiment not described in the SAR7 (3) Could the proposal affect nuclear safety in a way not previously
' evaluated in the SAR?
Anv antwer Yet y All answers No if
~
Most Technical Specifications (TS) require the 0nsite Review Group to (1) review all procedures and changes thereto that affect nuclear safety, all proposed tests
,, and experiments that affect nuclear safety, all proposed tests and experiments that affect nuclear safety, and all proposed changes to the facility that affect nuclear safety: and (2) to reconnend in writing to the Plant Superintendent approval or disapproval of these proposals.
10 CFR 50.59 no longer applies:
O Is an un-reviewed safety question invcived?
(1) 15 the probability of an occurrence or the consequences of an accident or malfunction of equipment important to safety previously -
evaluated in the SAR increased?
(2) Is the possibility for an accident or malfunction of a different type than any previously evaluated in the SAR created?
(3) Is the margin cf safety as defined in the basis for any Technical Seceification reduced?
.*e .
Most TS require the Onside Review Group to render deterininations in writing avith regard to whether or not the proposed change constitutes an un-reviewed safety cue' tion.
All answers No o Any answer Yes Most TS require the Offsite Review Group to review proposed changes to procedures. .
equipment or systems, and tests or experiments that involve an un-reviewaJ safety cuestie i.
V i Document the cnange. Include in these records 'g a written safety evaluation providing the bases -
for the determination that the change. test or sal to the esperiment does not involve an un-reviewed safety o u --
"" "" 4 4 l Author 124 tion received.l I Proceed with the enange Q I I
Most is require tne Of f site Review Group to review .
! the safety evaluations for changes to procedures.
' - - j equipment or systems, and tests or experiments completed under the provisions of 50.59 to verify that such actions did not constitute an un-reviewed safety Question.
l
(
I l
rule change would be avoided. Greater flexibility would exist for l l
implementation of the recommended criteria if~the TSIP recommended program was implemented as a Policy Statement rather than a rule. TSIP recognizes the industry's need for stability and would recommend that industry be given assurance of the NRC's dedication to implementation of the recommended program through issuance of a Commission Policy Statement. NRC resources for the implementation effort should also be identified.
The proposed criteria are clearly tied to the Design Basis Accident safety analyses. To evaluate the impact of accepting the recommended criteria, TSIP had to consider three areas. First, a most important consideration regarding the recommended criteria is their consistency with the present regulatory scheme governing Technical Specifications. The TSIP has carefully considered this matter and has concluded that the proposed criteria do isolate those systems, components and variables most important to safety consistent with' the current rule.
More specifically, the Statements of Consideration accompanying the current rule discussed the scope of Technical Specifications as incl'uding the following:
In the revised system, emphasis is placed on two general classes of technical matters: (1) those related to prevention of accidents, and (2) those related to mitigation of the consequences of accidents. By systematic analysis and evaluation of a particular facility, each applicant is required to identify at the construction permit stage, those items that are directly related to maintaining the integrity of the physical barriers
. designed to contain radioactivity. Such items are expected te be the subjects of Technical Specifications in the operating license. (33 Fed. Reg. 18610)
l 1
- j The proposed criteria, with their focus on immediacy and importance, are !
consistent with this emphasis. The first consideration, accident prevention, is captured by criterion (1) and to some extent, criterion (2) in that they address systems and process variables that alert the operator to a situation when accident initiation is more likely. The second consideration, mitigation of accident consequences, is clearly captured by criteria (2) and (3).
Table 2.1 summarizes the major ar.eas that would remain in Technical Specifications using the recommended program.
Items that would be removed from the Specifications, such as fire protection or primary coolant chemistry control, are not of immediate importance to safety. They provide essential controls and some level of control must be i
maintained over these programs, but the direct and immediate relationships discussed in the 1968 Statement of Conside.'ations and ALAB-531 is not evident.
Again, when 10 CFR 50.36 was proposed, the role of Limiting Conditions for Operations was discussed.
The ' minimum conditions for operation' would specify the lowest functional capabil_ity or performance levels necessary to assure safe operation of the facility.
The conditions generally would cover (1) the equipment or systems necessary to verify compliance with safety limits and (2) the required engineered safeguard systems.
(31 Fede'ral Register 10981) 1
TABLE 2.1 l
TECHNICAL SPECIFICATIONS REMAINING UNDER RECOMMENDED CRITERIA SAFETY LIMITS - (no change)
, LIMITED SAFETY SYSTEM SETTINGS - (no change)
LIMITING CONDITIONS FOR OPERATION - (and supporting Surveillance Requirements)
Reactor - Auxiliary Systems -
Shutdown Margin Service Water Modera;.or Temperature Coefficient Ultimate Heat Sink Moveable Control Assemblies Component Cooling Water Power Distribution Limits Special Test Exceptions Steam and Power conversion -
Reactor Coolant System - Safety Valves (Secondary)
Pressure Temperature Limits Auxiliary Feedwater Minimum Temperature for Criticality . Condensate Storage Appendix G Pressure Temperature Limits MSIV's (plant specific)
DNB Parameters Steam Generator Water Inventory Pressure Limits (Safety Limits)
Reactor Coolant Loops Radioactive Inventory -
Overpressure Protection (Normal, Liquid Holdup Tank Inventory Low Temperature) Gaseous Waste Tank Inventory Pressurizer Steam Generators Engineered Safety' RCS Leakage Features -
, Specific Activity ECCS Special Test Exceptions Containment Instrumentation - Electrical Power Reactor Protection System Systems -
Engineered Safety Feature AC. Sources Radiation Monitors AC Distribution DC Sources DC Distribution DESIGN FEATURES - (no change)
ADMINISTRATIVE CONTROLS -
Process to control and review changes.
Controls to assure adequate record keeping, review, and reporting Additional reporting requirements
The items which would be removed from Technical Specifications do not specify minimum levels, but provide added assurance above the minimums. They are not necessary to verify compliance with safety limits and thay are not engineered safety features. Therefore, the inclusion of these items in Technical Specifications is not appropriate. Removing them does not reduce the level of protection to below the level foreseen by the 1968 rule. This is not to say that some ite'ms will not be removed. On the contrary, Table 2.2 identifies a list of Limiting Conditions for Operations from the Standard Technical Specifications which may not be captured by the criteria but that currently require plant shutdown or power reduction. These LCOs should be evaluated against the subjective criteria of immediacy and importance to validate the proposed objective criteria.
The second area that was evaluated was the safety implication of removing any items from the Technical Specifications. As indicated in Section 5.1 on the evaluation process, TSIP assumed that less NRC control would imply a .
reduction in safety. This assumptior, is itself clearly conservative. In ALAB-531, the Appeal Board concluded: e "It bears repetition, however, that this should not be taken as reflecting a belief that the applicants are relieved of any obligation to take appropriate measures to live up to each of the commitments with respect to (spent fuel) pool operation which are set forth in the design report. For the reasons we have set forth, all we need or do decide here is that none of those commitments has been.shown to have such an immediate bearing upon the protection of the public health and safety that it must be mac' the subject of rigid operational limitation in the form of a Technical Specification. To the contrary, with regard to each commitment, the record affirmatively establishes that fulfillment of the requirements of 10 CFR 50.59 will provide ample safety protection." (9 NRC 279, 1979)
I However, even given the conservatism of that assumption, TSIP evaluated the subjective magnitude of this reduction in safety. First, TSIP noted that
~
removing any item from Technical Specifications does not alter the plant design. The assumption of decreasing safety comes from the reasoning that the removal of NRC pre-approval will lead to changes that allow degradation of the. system involved. To evaluate, subjectively, the potential impact, TSIP reviewed several sets of Technical Specifications against the criteria to determine if, in its own judgement, the systems that were most directly related to protecting the public health would remain in Technical Specifications. In each case, TSIP determined that the most important items were retained. TSIP also determined that the decrement in safety from removing some items, especially in light of the controls proposed in Section 2.2.2 of this report, was small.
Regardless of the magnitude of safety decrem e., any reduction should be justified by some improvement. This led TSIP to its third area of consideration. Were the safety reductions justified? Two possible justifications were evident. Allowing the removal of some items from Technical Specifications would result in either (1) resource improvements significant enough to override the safety concern, or (2) improvements in other areas which' would provide compensating safety benefits. The resource considerations are discussed more completely in Section 5.2 of this report but basically from the perspective of NRC resources there is no overriding resource consideration. Savings of approximately $2M and 20 professional i
staff years per year should result in NRR. This level would be reduced in i'
the short-term'by implementation costs of the changes. Long-term resource
considerations would certainly support a change but would not provide a compelling argument for NRC acti'on. Although the industry has shown a willingness to commit resources to this activity and appears to be highly motivated to pursue the improvement of Technical Specifications, they have not provided convincing resource arguments that lead to a conclusion that over-regulation is a.significant concern in this area. However, the motivation displayed suggests that the benefits anticipated by the. industry are considerable.
Because the resource arguments were not convincing, removal of some items for Technical Specifications was evaluated against the potential for improvements in other areas. By tying the reduction in Technical Sp.ecifications to improving the Bases and clarity of the remaining specifications, an overall improvement in safety should be realized. This.
1 compensating safety improvement provides the primary impetus for change.
A line-by-line review of the Technical Specifications against the existing safety analysis would be resource intensive. Development of additional analyses to expand the Bases would add more resource requirements. Even then there is no assurance that the NRC would approve the changes that deviated from the Standard Technical Specifications. Therefore, NRC needs to assure that the recommended criteria would be used as the basis for the evaluation and that reviews would be completed expeditiously and resources would be made available on a priority basis. In return, the utility desiring the changes would be required to conduct the review described above. The submittal would have to provide the expanded Bases, the improvements in clarity, and the disposition of, and controls for the requirements removed from the Technical Specifications.
o TABLE 2.2 SPECIFICATIONS WHICH REQUIRE REACTOR SHUTOOWN OR POWER LIMITATIONS.1 Power Distribution Heat Flux Hot Channel Factor Power Limitations RCS Flow Rate and Nuclear Power Limitations Enthalpy Rise Hot Channel Factor Reactor Coolant System Water Chemistry Shutdown in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Structural Integrity Restore or isolate prior to startup RCS Vents Shutdown in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> Pressurizer Heaters Shutdown in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> Engineered Safety Contain~ ment Structural Shutdown in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> Features Integrity Instrumentation- Remote Shutdown Shutdown in 7 days Accident-Monitoring Shutdown in 7 days 4
Containment Combustible Gas Control Shutdown in 30 days 1
l Electric Power Equipment Protective Devices De-energize crcuit 1
or shutdown i
Steam and Power Secondary Specific Activity Shutdown in 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> i conversion l
I 5
I Based on CP. AFT Former Tech-Spec LCO's providedsby AIF^as: applied to-Wolf' Creek
)
Technical Specti; cations.
4
21-2.2.2. Increased Attention to 10 CFR 50.59 l
The NRC should give increased attention to the review of changes made by licensees using the 10 CFR 50.59 process.
In Section 4.1, TSIP discusses the reluctance of NRR reviewers to impose controls on a licensee by means other than the Technical Specifications.
While the regulations now require updating of the FSAR, neither NRR project managers nor technical reviewers have been given formal responsibility to review these updates. Using 50.59, licensees make hundreds of changes a year which receive no NRR review or attention. Only changes which licensees conclude involve an unreviewed safety question or a Technical Specification amendment currently receive NRR review and attention. Also, while most 50.59 changes are appropriate, NRR personnel are aware that occasionally licensees have overstepped the intended use of 50.59 by not reviewing changes that may have involved an unreviewed safety question or by making an j inadequate safety evaluation. To an NRR reviewer, this combination-of factors implies that a licensee may later change an FSAR commitment upon which the reviewer relied in the SER at the time the plant was licensed. Therefore, the reviewer mistrusts the 50.59 process and relies instead on Technical Specifications to validate the commitment.
NRC Inspectors do (1) review the lists of changes made by licensees under 50.59 each year; (2) audit the licensee's procedures for complying with 50.59; and (3) review the documentation of several of a licensee's 50.59
o changes each year. However, NRR personnel are rarely involved in the NRC review effort for 50.59 changes even though it is the NRR personnel who are the most familiar with the licensing basis for the plant.
To alleviate this mistrust and provide appropriate NRC attention to those conditions which would be removed from the Technical Specification using the recommended criteria, TSIP recommends that the NRC give increased attention to review of 50.59 changes. First, NRR project managers should be formally and specifically charged with responsibility for review of FSAR updates. The project manager would be responsible for identifying changes which potentially overstep the intent of 50.59. The project manager would consult with I&E and regional personnel and with NRR management and technical reviewers as needed
, and initiate further action such as formal NRR reviews when necessary.
Second, NRR should set up a procedure with Regional Offices to give increased attention'by including joint NRR, IE, Regional reviews of selected issues.
8 The proposed solution would entail an increase in emphasis on 10 CFR 50.59.
Therefore, NRR, IE, and Regional personnel would have to be informed of and provided some level of training in the role of rules, license conditions, Technical Specifications and FSAR commitments. This training should include descriptions of the roles of project managers, technical reviewers, regional-based, and resident inspectors. It should also include a comparison of the relative level of effort to be devoted to Techaical Specifications compared to other areas, such as 50.59 and would possibly require revisions to the IE Inspection Manual.
2.2.3. REVISIONS TO TECHNICAL SPECIFICATION BASES AND CONTENT NRC should review and revise the Standard Technical Specifications to correct human factors and other technical weaknesses through a program of technical assistance and dedicated in-house resources.
2.2.3.1 Short Term Improvements In addition to the programmatic changes discussed above, there are three
- specific actions that should be taken to improve the Technical Specifications that fall.within the recommended criteria. These are
(1) Provide more complete and meaningful Bases. Virtually all' organizations interviewed by TSIP indicated a need for well documented and technically supported Bases. Perceived benefits included a potential for safety enhancement through a reduction of unnecessary shutdowns, improved equipment testing requirements and action statements, and better use of resources by reducing the many Technical Specification interpretation problems. These problems occur daily in areas of compliance, enforcement, and amendment. A potential for reduction in occupational exposure exists
! if unnecessary test requirements are eliminated.
The work sponsored by TSIP to update the containment Systems Technical Specification Bases (Kripps, September 1985) is an example of an effort to collect already available information for incorporation into the Bases. Follow-en work for formal review by the Owners Groups and for b
o additional technical ass.istance to update other sections of the Technical Specifications should proceed in concert with Owners Group efforts.
Owners Group initiatives and submittals to improve Bases should be 3 encouraged. The AIF Subcommittee on Technical Specification' Improvements states that all four Owners Groups are planning to utilize probabilistic methodology on specific Technical Specification requirements. This area provides significant potential improvement and is discussed further in Section 4.3.1 below.
t (2) Correct problems with clarity of the Technical Specifications. There are several instances in which discussions have centered on what a requirement means as discussed in Section 4.3.2. The resources wasted l on interpreting these requirements could be saved, but a line-by-line cleanup of t.he Technical Specifications would be required. The AIF I
Subcommittee has praposed that a Writer's Guide be prepared (probably by the ANS 58.4 working group) to provide clear guidance for revising the specifications. This approach has been used in the development of upgrading Emergency Operating Procedures (EOP), and there may be efficiencies realized by using that effort as a basis. TSIP recommends that this be pursued with AIF and ANS and that the Technical Specifications be reviewed for clarity and useability.
i (3) Examine the operability definition and requirements. In 1980,. the NRC required a general definition of operability be included in all Technical i Specifications (Eisenbut, April 1980). Coupled with the general requirements that a system that cannot be proven operable within the
scope of the Technical Specifications be declared inoperable and the requirement that, barring a separate action statement, plant shutdown should he undertaken when equipment operability problems persist, the
, application has been extremely conservative. This has led to shutdowns to " correct" situations where slight system degradation has been interpreted as syst,m inoperability (Gallagher, July 1985). Clearly, there must be a point at which a system is declared inoperable. However, given the lack of Bases for most Surveillance Requirements and LCOs noted above, a strict interpretation of the operability definition does t
not seem warranted.
The AIF Subcommittee has been developing a proposal for improving the operability definition. Because TSIP has not had an opportunity to review the proposal in depth, TSIP is unable to provide a recommendation on its merits. However, TSIP recommends that the OPERABILITY definition and the associated general requirements, such as Technical Specification 3.0.3, be examined for improvement. In the meantime, TSIP would recommend that, for surveillance requirements and LCOs that are not clearly related to the safety analyses, discretion be used. Licensees should still observe the Technical Specification requirements strictly and request relief as early as possible.
Two cases should specifically be considered, First, when an allowed outage time is likely to be exceeded by some small amount but there is a high probability that the system will be returned to service shortly beyond the A0T, there should be discretion in allowing continued operation. Second, when a surveillance test has been missed but the system can be shown to be operable by analysis or other means, can be tested by tests other than those
l required by the surveillance requirement, or it can be demonstrated that the safety function assumed in the accident analysis is not threatened, then one-time relief should be granted. The more tightly coupled the A0Ts and Surveillance Requirements are to the safety analysis (such as for Reactor Protection System testing consistent with WCAP 10271), the less discretion needs to be used in evaluating the licensees request for relief.
The AIF Subcommittee has also developed suggestions for extending the 18 month fuel cycle requirements, for deleting cycle-dependent variables from the Technical Specifications and for revising the general Technical i
Specification on changing modes (T.S. 3.0.4). TSIP has not had an opportunity to review each of these proposals in detail. However, if a detailed review of the Technical Specifications is to be undertaken, these areas should also be addressed. Therefore, an NRC position on the general acceptability for each AIF proposal should be developed.
l The NRC should give priority to cooperative efforts by owners groups or AIF/ANS, such as proposed revisions to the Standard Technical Specifications or development of a writers guide. NRC review methods should recognize generic work and avoid duplicative reviews of individual submittals that
! closely follow approved generic programs. The NRC should reject programs that propose applying the criteria for Technical Specifications without providing the improvements to the remaining Technical Specifications.
2.2.3.2 Long Term Improvements l Since the safety analysis'in post cases did not address system testing or action statements in the event of system degradation, it will be impossible
= . _ . _ . - _ _ - - - . - - . _ _ _ _ . _ - - _ - - _ _ - = _ _ . - . -
~
I
< l to derive complete Bases from the safety analyses. Therefore some will have to be inferred and the linkage documented, and some will have to be developed.
j TSIP recommends that a comprehensive program to improve the Bases be established.
7 i
j As the efforts to improve'the Bases progress several problems may be identified. This is an ideal time to correct these problems. Where.
I
! evaluation of the Bases indicates inappropriate action statements, the i statement could be revised to more closely follow the overall safety analyses.
A second area to evaluate when reviewing Bases is the appropriateness of test '
i methods. The test selected should ensure that the specific safety function j assumed in the safety analysis is being verified. Scrutiny of the Bases'and closer association with the LCOs and LSSSs may identify inappropriate test i methods.
I 6 i
Finally, the evaluation of Bases could identify inappropriate test frequencies.
i Since for the containment system Technical Specifications reviewed, few I
j Surveillance Requirements could be linked diractly to the safety analysis
{ through the Bases, there appears to be considerable room for improvement in ~
j this area. Probabilistic methodologies discussed in Section 2.2.4 could provide additional information for improving this area.
i I
i l The impact on Technical Specifications should be considered by the NRC staff 1
{ in evaluating new safety information. Forthcoming Commission decisions on j new reactor safety information, particularly in areas of piping analysis, t
j seismic margins, source term and ECCS models should address the impact on l
l
Technical Specifications. The new information should provide an overall improvementoto the Technical Specifications through justifying, as appropriate, relaxati.ons, strengthening, and changes that correct some existing problems and errors. Incorporation of improvements derived from new information on seismic margins and piping analysis could proceed in the near future as Commission policy in these areas is aavancing rapidly. Improvements based on new source term information should be studied now, but actual implementation must be paced by the development of final Commission policy in the area.
Improvements in ECCS Technical Specifications would be paced by rulemaking activities in this area now underway.
1 4
i 1 2.2.4 Use of Probabilistic Risk Assessment Methodology to Improve Technical Specifications ,
The NRC should encourage the continued development and application of probabilistic risk assessment methods to address Tecnnical Specification
- requirements.
1 t
j There has been an increasing trend toward the use of risk-based evaluations to assist review of Technical Specification requirements (and their associated 4
Bases). Westinghouse and General Electric Owners Group analysis of Reactor Protection Systems (WCAP 10271 and NED0 30855;) and Commonwealth Edison's Byron LCO Relaxation Program are recent examples. At this time, there is not
, a standard methodology or review method for using probabilistic risk assessment methods to assist determination of Techn.ical Specification requirements. As a result, reviews have taken quite a long time and were l i
t l
,_ , , - , - , , - - - - - - - - - , - , - - - - , - - - , - - - - - , > - - - - - , , , ~ - - , --c-- -,- ---,-
quite extensive. Two of the most difficult problems with the use of PRA are the lack of a cumulative outage consideration and lack of acceptance criteria (relative change in risk or absolute risk level). Of course, other uncertainties also exist involved with the modeling, data and assumptions.
Nonetheless, the insights gained from these recent evaluations have been and are being used to successfully alter Technical Specification requirements where-justified.
Owners Group efforts should be encouraged. The AIF Subcommittee on Technical Specification Improvements stated that all four Owners Groups are planning to utilize probabilistic methodology on specific Technical-Specification requirements. Efforts to effectively utilize NR'C review resources should continue. These efforts should be coordinated with the continuing research by industry and NRC to develop _ methods and Procedures to Evaluate Technical Specifications (PETS). Coordination should be the responsibility of the Technical Specification Coordination Branch (TSCB) with assistance from the Reliability and Risk Assessment Branch and the necessary NRR technical branches.
The Office of Regulatory Research (RES) PETS program to develcp methods to evaluate the Bases and extensions of allowed outage times (A0T) and surveillance test intervals (STI) is very close to completion. Industry participation has been active and should continue based on its stated interest. Efforts to finalize and use the procedures should proceed in the next year. As soon as practical after completing the A0T/STI work, risk
a a - ---LJ-j.
t assessment methods should be evaluated that could assist the industry and NRC in the determination and improvernent of Technical Specification Action i
Statements.
Following the allowed outage time and surveillance interval work, it is 4
recommended that the work related to cumulative outage issues be completed by RES and the results provided to NRR with a recommended course of dction.
Like all other PETS work, review and comment by the public should be encouraged. At this time, the AIF has indicated that there is no clear benefit to a cumulative outage time. This issue must be considered if probabilistic risk assessment is used to increase allowed outage times for equipment.
The values and impacts of cumulative outage requirements should be considered by NRR as necessary in accordance with NRR Office Letter Number 40 (Denton, March 1983). Every effort should be made to involve the industry in each phase of this research. It is recommended that the AIF Risk-Based Methodology Working Group be used as a contact for this review. A related issue of j equipment configuration management (combined equipment Technical Specifications)
- was recently proposed as a new generic issue. This proposal would provide the control of equipment / components in the same accident sequence cut set, i
We recommend that prioritization of this proposal proceed in accordance with
- NRR Office Letter Number 40.
In the long term TSIP supports programs that would investigate alternatives to prescriptive Technical Specification test and maintenance requirements, i
i
31-4 I
The Office of Research program, Operational Safety Reliability Research Project, is investigating such an alternative. NRR should continue to evaluate this effort's trial application and participate in the peer review in FY 1986.
3 Impacts of Implementing Recommendations 3.1 Safety As discussed in Section 2.2.1 TSIP concludes the overall safety impact of the recommended program is oositive. This is based upon the safety enhancements expected from the technical improvements and Bases development which are I
conditions for implementation of the program. The negative component is i
based upon the reduction in NRC control over those requirements removed from i the Technical Specifications; that is, it is assumed that a reduction in NRC control has an adverse safety impact. This effect is offset however, by the fact ~that only those requirements not meeting the criteria of " immediacy and importance" may be removed from the Technical Specifications, and these same requirements will now be covered by the 10 CFR 50.59 control process will receive increased NRC attention.
i 3.2 Resource Impacts 3.2.1 Near Term (1-2 years) l The primary contributor to the resource. impact in the near term is the effort to reorganize, clarify and provide Bases for those items which remain as
i Technical Specifications. The effort to organize and clarify the specifications will be facilitated by the industry initiative to develop a
" Writer's Guide" for Technical Specifications. It is still expected that, as a minimum, a staff effort similar to the " proof and review" effort expended on recently issued Technical Specifications will be necessary for each plant that chooses to revise its Technical Specifications. The magnitude of this i impact could be minimized by developing model bases for the Standard Technical
- Specifications. Although this effort is estimated to require $2M-$4M (exclusive of new analysis beyond that already available) plus some in-house j support, it would result in less in-house resource expenditure for the reviews.
The effoit increases with the number of ?,echnical changes proposed. Total impact therefore depends upon the number of plants which elect to modify their specifications and the extent of the technical modifications. There ,
will also be the associated burden of a large number of license amendment packages.
The other near-term programs will also have resource impacts, but these impacts will not be of the' magnitude associated with the actual revisions.
These efforts include developing the Policy Statement, modifying some regulations to allow full application of the recommended criteria, developing a guidance document for Technical Specification content, refining the l OPERABILITY definition, and encouraging the use of PRA for technical enhancement. Also included is the effort required to inform the NRC Staff of the roles of regulations, license conditions, Technical Specifications and FSAR commitments and of the roles of project managers, technical reviewers, and region-based and resident inspectors.
i l
.- - - . - -,n- ,, , - - -
3.2.2 Long Term l The long-term benefits a'aing depend primarily on the number of licensees which chooses to revise their Technical Specifications. Applying the recommendations, approximately 1/3 to 1/2 of the existing LCOs would be moved t out of the Technical Specifications. It is expected that there will be a comparable reduction in license amendment requests in the-future. For the i
Regional Staffs, the problems of interpreting the Technical Specifications during inspections and evaluations of operations would be significantly reduced. The improved Bases and clearer wording'would eliminate many of the protracted discussions described to TSIP by the regional personnel.
However, for those items removed from Technical Specificatons, the burden would shift to the NRC to prove that the. licensee's review had been-inadequate. This has the potential for adding significant resource burdens
~
to the regional staffs. Therefore, shifting the emphasis and some of the resources currently used by NRR and the Regions from evaluation of Technical Specification compliance to review of the li'censee's programs for control under 50.59 will provide confidence in the process. The net effect would be neutral over the near-term, and as confidence.is gained, resources could be channeled into other areas for a net savings.
Another potential benefit is an increase in plant availability due to a-i reduction in downtime associated with unnecessary or overly conservative specifications. -This benefit is also the result of the clarification and Bases development efforts. Based upon the TSIP review of Technical i
l
J Specification outages in 1984 which appearedJto be unnecessary, and' assuming
$300K/ day replacement power cost, the potential exists for a per plant savings of $20K to $200K per year (Beckham, August 1985). The BWR Owners Group estimated a potential savings of 1280 to 6240 megawatt-hours per 1000Mwe
- plant per year would result if inadvertent reactor trips from unnecessary reactor protection system trips were eliminated (Sullivan, May 1985). At 3C/kWh, this is equivalent to $45k to $200k per year per 1000 Mwe plant.
4
, 3.3. Administration TSIP has recommended that 10 CFR150.36 be retained in its current form. A j Policy Statement would require significantly less time to develop than a new rule. Since the overall program is voluntary for licensees, a Policy Statement would be sufficient to provide the guidance necessary to direct NRC
, staff activities. Application of the criteria to current LCOs would be constrained by other regulations. Table 3.1 lists substantive regulations currently affecting Technical Specifications. ' Relatively minor changes to
- these rules should allow the flexibility necessary to implement Technical Specification improvements. The changes should not allow deleting items from Technical Specifications unless the compensating improvements are made.
Changes to the regulations that are intended to reflect new safety information should not be confused with the changes necessary to allow a licensee to improve its Technical Specifications.
Enforcement activities could become more difficult under the recommended program. Under the present scheme calling for prior approval of Technical l Specifications changes, the only enforcement issue is-literal compliance with.
\
4 -
r- --- n- . - . - ,,m. .7_n- w -,n. . , , , -my.--e-, ~,y _ --mm.,-, ,-
TABLE 3.1 -
SUBSTANTIVE REGutATIONS IN TlfLF 10 INVOLVING TECHNICAf 5PECIfTCATION5 00R POWfR RFACIORS Primary Regulation General Description Disposition Under Proposed Solutions 10 CFR 50.36 Describes Technical Specifications (IS) '"
10 CFR 50.36a retain Extends 15 to effluents modify 10 CFR 50.59 Changes to 15 must he by license amendment retain Sicondary Regulations 10 CFR Part 2 Appendix C Specific references to TS to determine severity modify (Enforcement Policy)*
. violations 10 CFR 50.46 ECCS systems modify 10 CFR 50.48 Fire protection modify 10 CFR 50.54 Minimal shift requirements, cold shutdown - modify refueling, emergency action 10 CFR 50.55a inservice inspection modify 10 CFR 50.71 Record retention 10 CFR 50.72 and 73 retain Notifications retain 10 CFR Part.50 Appendix I Effluents 10 CFR Part 50 Appendix J Leak testing modify I" 10 CFR Part 50 Appendix K modify Maximum peaking factor modify 10 CFR Part 50 Appendix H Pressure-temperature limits 10 CFR Part 50 Appendix R modify Fire Protection (reference to Standard TS) modify 10 CFR 55.?2 Tests for operators 10 CFR Part 55 Appendix A retain Re qualification program retain 10 CFR 70.32 Special Nuclear Material (physical security plan) modify
'Stric,tly speaking this is not a regulation but rather Commission policy.
t
- n 5
the Technical Specification. Under the recommended program the licensee j could unilaterally make changes to many areas previously requiring prior approval. If the NRC concluded that the licensee's 10 CFR 50.59 review was inadequate and that enforcement action was appropriate, the NRC would have the burden of proving the licensee wrong.
The' magnitude of the potential impact, however, is difficult to gage If the NRC implements an appropriate program of oversight of 10 CFR 50.59, enforcement impact could be minimal. The. proposed Commission Policy Statement should identify that the NRC will pay increased attention to 10 CFR 50.59 reviews. Also, enforcement would benefit from the upgraded Technical Specifications because less interpretation by inspectors and reviewers would 1
i be required.
4 PROBLEM IDENTIFICATION This section (1) describes TSIP's approach to identifying Technical Specifications problems and (2) discusses the problems identified by TSIP and the significance of those problems. TSIP's Interim Report provides additional details on the problem identification stage of the project.
1 TSIP gathered information from various sources within and outside the.NRC:
j Within NRC:
Technical Specification Advisory Group NUREG-1024, " Technical Specificationst. Enhancing The Safety
- Impact"
i
- Previously published Proposed Rule Chang'e to Split the Technical Specifications (47 FR 61, March 1982) ,
Technical Review Branches Regional and Resident Inspectors !
i Technical Specification Review Group Division of Licensing Project Managers
, licensing amendment survey Regulatory documents: 10 CFR 50, ALAB-531, past studies.
Office of' Inspection &. Enforcement Technical Assistance:
Bases Adequacy Impact on Operating Reactors - 1984 Data Risk Significance of Technical Specifications Effects of new information (source term, seismic margins, i
and leak before break).
Outside NRC: ;
AIF Technical Specification Improvement Subcommittee 1
Plant operations staff interviews Regulatory Effectiveness Meetings (RES~ program)1 Department of Energy (FFTF)
Westinghouse Hanford (FFTF) 9
-l 1
See Table 4.'1
. - . . ~ . . . _ _ ~ . _ - - . _ , . _ . _ _ _ _ _ , , _ ._ . _ , - - , . . _ . _ _ _ . . , . . - . - , - . . .
1 j
TABLE 4.1 PLANT OPERATIONS STAFF INTERVIEWS BY TSIP i
J 4
St. Lucie Limerick l Sequoyah Peach Bottom t Oconee ANO-1 La Salle ANO-2 Dresden Trojan i Pilgrim . Zion r
D. C. Cook l
l [
1 UTILITY / VENDOR OFFICE VISITS i i
i l Tennessee Valley Authority ,
I '
i Duke Power f- Florida Power and Light i
Georgia Power l Westinghouse i i
4 4
l 1
Office of Research Regulatory Effectiveness program 4
_ , , .,,r, . . _ , . , . . . . - , - - - . . - . _ , . _ , . - . . . , , , , . , ._.-.,.,-,.._-,,.-.--_-...,_,,-_-,,,,..,,...,-,...,-,.s-. , ~ - ~ - ,
39 The information gathered by TSIP was evaluated and resulted in the identific~ation of three root problems associated with Technical Specifications.
- 1. Reluctance of the NRC staff to use tools other than Technical Specifications for establishing regulatory requirements.
- 2. Lack of well defined criteria for what requirements should be included in Technical Specifications.
- 3. Human factors and other technical weaknesses of the Technical Specifications.
TSIP concludes that these concerns do not pose an acute safety problem for operating power reactors. TSIP's alternative solutions evaluation in Section 5 will discuss potential safety impacts in more detail.
4.1 Use of Alternate Tools The reluctance of the NRC staff to use regulatory tools other than the Technical Specifications is generally attributable to one of two causes:
a limited understanding of the roles of the different tools available or a lack of confidence in the effectiveness of alternative tools. These problems lead to the general belief that concerns must be addressed in the Technical Specifications to ensure compliance and, thereby, assure the public health and safety.
. . _ - _ = .
i This belief, along with the lack of criteria delineating what items should be 1
included in Technical Specifications has led to the inclusion of too broad a range of items in the Technical Specifications.
t 4.1.1 Role of Regulatory Tools I
l When the NRC licenses a nuclear plant, the safety analyses form part of the i
- basis for the NRC's conclusion that the plant can be operated safely. After i licensing there is a continuing regulatory need to ensure that the plant operates in accordance with the assumptions in the safety analyses. This link between licensing safety analyses and plant operation forms a part of the continuing basis for the NRC's conclusion that the plant should be allowed i to operate. This judgement is the responsibility of the NRC. The Technical i Specifications are one tool which the NRC uses to provide this link between
] licensing analyses and safe plant operation. Other tools are the regulations, the FSAR/SER, plant procedures required by regulation or license condition, j and programmatic documents required by regulations such as the. licensee's
! Quality Assurance Program, Security Plan, and Emergency Plan.
i 1
i j The regulations are a set of rules which licensees must obey even if those rules are not duplicated in the Technical Specifications. The regulations I
(1) require that there be Technical Specifications; (2) describe in broad categories what should be covered by Technical. Specifications; (3) state the l NRC procedures for control of Technical Specific'ation changes; (4) require
- that there be an FSAR which describes the facility, and limits on its operation; (5) require the FSAR be keot up-to-date; and (6) state the procedures for l control of changes to the FSAR.
}
l
e 41 t
4 When a plant is' licensed, the NRC reviews the FSAR and prepares an SER j' approving the facility and providing the basis for that approval. Often the SER includes conditions which must be met to make approval valid. These {
conditions can be covered in the license or the Technical. Specifications l
} which are part of the license. Some of these conditions can be met by l
- revising the FSAR.
l The Technical Specifications are derived from the FSAR safety analyses (approved by the SER) and define the operating safety envelope within which a plant should stay. The Technical Specific ~ations also state what is to be i done when the plant cperates outside those bounds. The Bases explain why the Technical Specification limitations are what they are. These Bases support the Technical Specifications and are attached to the specifications, but they j are not legally a part of the Technical Specifications. The NRC may issue .
1 civil penalties for violations of the Technical Specifications or the I regulations as described.in the NRC's Enforcement Policy (10 CFR 2, Appendix C).
F 1
Licensees develop procedures, which are referenced by regulations and by the i
- Technical Specifications, which describe in detail actions to follow in i
operation of the plant. These procedures are designed to keep the plant ,
I within the Technical Specification limitations. Procedures are the instrument l with which operators deal most frequently.
- There are a number of other programmatic documents which are referenced by j the regulations and/or Technical Specifications such as the Security Plan, i
i i
I Emergency Plan, and Offsite Dose Calculation Manual (00CM). In most cases, ;
i various procedures were set up by the regulations or Technical Specifications to control changes to these programmatic documents.
i
. These above descriptions explain the relationship of the various regulatory 1
documents or." tools." In examining these relationships, TSIP concluded that-FSAR commitments are enforceable through 10 CFR 50.59 and that programmatic documents required by regulation or Technical Specifications are also
- enforceable. However, TSIP found a considerable amount of variation throughout the NRC in the knowledge of 10 CFR 50.59 procedures, and 4
! consequently, on the trust placed in the process for ensuring control.
i l Because of this, TSIP found that many of the NRC staff (headquarters and 1
j regional) hold the philosophy that if it is not in the regulations or the i Technical Specifications, the NRC can not ensure that the licensee will do 4
i t. Certainly NRR Office Letter No. 34 Revision 1 " Utility Commitments" (Denton, July 1981) encourages this philosophy.
1 i
j Consequently, Technical Specifications have been used for matters that deal i
with a wide range of safety importance. TSIP concludes that many of these matters are inappropriate for inclusion in the Technical Specifications and could be effectively and more efficiently controlled by other mechanisms.
4.2 Lack of Criteria TSIP identified a lack of criteria delineating what limitations and conditions 4
j should be included in the Techaical Specifications. The absence of clear '
i l
~_ . - - - . - _ - _ - . _ - - - - . . - - - . _ _.-_. - - - _ _ - - . _ _ -
43-i f
i criteria for what should be in Technical Specifications may have grown out.of
] the lack of understanding of the role of regulatory tools described above.
1 5 ;
l Probably the most definitive guidance as to how Technical Specifications are i i <
{ to be used is NRR Office Letter No. 34. This document instructs the NRC i !
staff to use the Technical Specifications to provide legally binding <
! l 1 requiremerts in the license for any limitation or design detail which is i !
] significant in the formation of the safety evaluation. i
! NRR Office Letter No. 34 states: l i
4
}
- "If the commitment is of such importance that no change should be made without prior staff review and approval, j , it should be reflected in the technical specifications i or as conditions to the license."
i This guidance leaves the decision about what should go into the Technical -
! Specifications primarily in the hands of the NRR technical ~ reviewer.. This t i guidance does not provide criteria to the reviewers or NRC management about the~ depth of the analyses regarding which items must be included as Technical j Specifications or what systems are important enough from a safety standpoint [
to be Technical Specification items. As a result, there is confusion over
{
j wnat requirements warrant inclusion in the Technical Specifications, and I
therefore, Technical Specifications are imposed for items which vary greatly I
- in their importance to public health and safety.
It is difficult to identify specifically the significance of this lack of j principal purpose and criteria in terms of plant safety implications or I
u _ __ _ __ . _ . _ _ _ _ _ _ . _ _ _ . _ _ _ . _ _ _ _ . _ _ _
resource burdens on the NRC and licensees. However, the confusion and the resulting inclusion of items of lesser safety importance in the Technical Specifications does tend to divert attention (including that of the operators) from the more important safety items in the Technical Specifications. The presence of the items of lesser importance in the Technical Specifications also results in incremental resource costs to licensees and the NRC in administration and the change process.
4.3 Human Factors and Other Technical Weaknesses TSIP has concluded that another important problem with Technical Specifications is that the NRC has not given sufficient consideration to making the Technical i
Specifications easily usable from an operations standpoint. Past emphasis i
has been on licensing and enforcement, with little regard for operations, j Without a clearly defined statement of purpose, it is difficult to identify the primary user for whom the Technical Specifications are written. This problem of emphasis is fundamental to the subsequent problems of clarity and Bases which are discussed in the following sections.
4.3.1 Bases The Technical Specification Bases are inadequate. Many are non-existent, few are complete. The Bases in general are not unique for each specification, do not reference analyses to explain the determination of the 1.C0, do not i
discuss mode applicability, do not discuss how action requirements relieve j safety concerns, and do not support Mst requirements, frequencies or allowed I
outage time:. These problems with Bases have been identified by the Technical Specifications Advisory Group, Technical Specification Task Group (NUREG-1024), l
i i
l staff reviewers, regional staff, resident inspectors, plant operators and maintenance personnel, utility compliance engineers, and public interest i
i groups.
1 I
! A significant impact of this problem concerns the Technical Specification change process. Neither the licensee nor the reviewer has the necessary l
background information readily available to facilitate changing a Technical Specification. If it is uncertain why a Technical Specification is as it is, it is very difficult to justify a change to it. This contributes to the log-jam of change requests, the reluctance of staff reviewers to approve changes, and consequently, to the negative attitudes expressed by licensees
! and operators toward change requests.
I A second impact of the inadequacy of the Bases is aggravation of interpretation i problems. The Bases should provide information which clarifies the safety i
intent of the specifications. When they do not, misinterpretation can lead to inappropriate actions which may result in a less safe plant condition or l an unnecessary shutdown. The same concern applies to the preparation of test procedures. Without clear Bases to define the safety intent of a specification, l
i it is left to the judgement of the procedure writer or test personnel as to l
how a test should be performed. The result may be an inappropriate or less
]
efficient test. Finally, enforcement becomes more arbitrary when interpreta- ,
tions must be made without supporting Bases.
t From a safety standpoint, lack of a documented basis for requirements is most significant in the area of equipment test frequency, action statements, test types and allowed outage times. AlthoucS not an acute safety concern, the potential exists for safety enhancement since (1) maintenance is in many
i 2 l
4 instances the dominant contributor to system unavailability; (2) test and maintenance errors are a significant cause of inadvertent reactor trips; (3) Action Statements may be directing plants to less safe conditions; and (4) equipment designs do not always include appropriate features to facilitate
( proper testing.
i 1
4
[ It should also be recognized that a body of new information upon which to 1
4 base nuclear plant safety analyses and practices has been developing from l reactor operating experience and NRC and industry research programs. The J
- most important areas are source term applications, piping failure mechanisms j and seismic margin. Recognition and incorporation of this new information is expected to result' in both safety and administrative benefits, particularly-a l j for the containment systems, piping support requirements, and component and J system operability requirements. Moreover, there are indications that several
! of the current Technical Specifications in these areas may involve a negative safety impact and others are, at best, benignly erroneous because they are ;
l based on outdated information. As part of the TSIP technical assistance l effort, a contractor has surveyed Technical Specifications from the four i l! reactor vendors and found for the sections on instrumentation, reactor coolant j system, emergency core cooling system, and containment systems that.as many as 38% of these Technical Specifications may need revision (Phung, August 1985).
i I L
{ 4.3.1.1 The Bases Problems and Their Significance t 4 ,
[
I ,
l 2
An inspection of the containment systems sections of all vendor Technical l 4 Specifications indicates that the Bases for most Technical Specifications are
=
l a .- - - -- - - - . . - - - - - -
47 almost entirely absent, the action statements are different from vendor to vendor, some surveillance tests do not totally test the system safety function, some LCOs are not based on the safety analyses, ahd some Action Statements make compliance impossible.
The following shows the extent to which these problems were found in a review of containment systems Bases.
BASES AREA % OF BASES COMPLETE Safety function / definition 95%
How safety concern is aggravated by violation i
of LCO 20%
Reference to analyses used to derive specification values 0%
Explanation of mode applicability 0%
Basis for action statement 5%
I Surveillance requirements 10%
Some of the missing Bases can most likely be found in the FSAR and other documents while others cannot. This is a consequence of the safety analysis primarily supporting the design and construction of the plant rather than its operation and maintenance. The Technical Specifications relating to.these areas are largely based on judgement backed up by manufacturer's' recommendations, standards, and some analysis which, in most cases, is not documented. It is these areas, operation and testing, where most of the concerns identified to TSIP were focused. While missing Bases are a concern by themselves, concerns
--w,-e . wy __ _
_-,9 .p - -. - ,. , ,.-.-- c _. - - _m-.,.-%7__ ,
also exist with specific allowed outage times, test types, test intervals and Action Statements, and therefore, with the associated Bases. For resolution of these Bases problems, additional analysis and review would be necessary.
One specific concern was that Technical Specifications were unnecessarily shutting down plants. A review of all shutdowns required by Technical Specifications in 1984 indicates that some shutdowns occurred when continued operation probably could have been justified. These shutdowns did not result in significant outage time. However, most of the shutdowns were judged to be necessary. The average unavailability per plant (industry-wide) from shutdowns required by Technical Specifications was about 1.5 percent for 1984. Total average unavailability per plant (industry-wide) for 1984 was 37
- percent.
To better understand the significance of the problems with allowed outage times and surveillance test' intervals, TSIP requested the assistance of the Office of Regulatory Research. Using.a plant specific PRA and the plant's Technical Specifications, risk estimates were made for the specific allowed outage times and surveillance intervals. The study estimates an extremely wide range (6 to 7 orders of magnitude) of allowed risk for various degraded modes of operation and test intervals. In addition, about half of the requirements had minimal risk impact (Samanta, August 1985). This is not a surprise, (although the magnitude of variance may be higher than anticipated),
since the Technical Specifications were never intended, by themselves, to regulate to a certain level of risk or system reliability. The current s
requirements for~ Technical Specifications are based pr%arily on the single
49 failure criterion and deterministic safety analysis assumptions. For the most part, there is no attempt to maintain operability of systems or components -<
that are part of the same probabilistic risk assessment accident sequence cutset. Nor is there a. specification on the total cumulative time that a component, train or system can be out-of-service over a period of time.
4.3.2 Clarity The Technical Specifications often lack clarity. The organization is poor, the use of footnotes is confusing, and the wording needs improvement. A common complaint of operations personnel is that a single component or system may appear in several Technical Speci'fications with d.iffering requirements but without cross referencing. Testing and maintenance personnel indicated a difficulty in ascertaining what tests are necessary and sufficient because of poor wording in the Technical Specifications. The use of adverbs such as "immediately" and " continuously" causes enforcement hassles. Some requirements are only applicable-in one mode, but no direction is given for what requirements exist in other modes. These complaints were universal among operators and operations staffs at the plants. A few who had lived with these problems for a while were reluctant to propose a change for fear of making the problem worse. (Clearly, many of these problems relate to the inadequacies of the Bases.)
l
, i l
1 The consequences of this problem are uncertainty, and sometimes frustration,
! on the part of operations and testing perso;.nel. This can result in j
inappropriate actions which may take the plant to a less safe condition, cause excessive testing, or cause unnecessary plant shutdown. Licensees i
j often develop numerous interpretation documents to state their view as to what a specific Technical Specification-means. Resident inspectors are called
- upon to assist in understanding Technical Specifications, and the meaning of Technical Specifications can change as resident inspectors change.
i l'
Unclear Technical Specifications also create problems in the implementation j of the agency's enforcement program. Technical Specifications often form the basis for the issuance of proposed civil penalties in accordance with the NRC's Enforcement Policy. Substantial resources are often committed to 1 l determine whether or not a violation occurred. To the extent a licensee is I unsure of a Technical Specification, compliance becomes frustrating. i Licensees generally indicated a strong desire to meet Commission requirements.
f To the extent Commission requirements, in this case Technical Specifications, i
- are unclear, the agency is losing the benefit of a positive attitude by ,
j licensees toward compliance. This weakens the enforcement program.
l
- T, SIP identified a number of problems associated with the Technical Specification definition of OPERABILITY. These problems are included but are ;
strongly associated with the lack of Technical Specification Bases. First, the current definition of OPERABILITY requires 1.icensees to declare'a component inoperable if a surveillance requirement associated with that component is missed. If in fact there is no other rearon to believe that the component is inoperable (unable to perform its intended function), then missing a surveillance does not make the component inoperable.
51- l l
l Based on a review of all LERs submitted in 1984 to report deviations from Technical Specifications, over 100 missed surveillance tests were discovered for U.S. reactors as a whole. Usually, the licensee is able to perform the test before the applicable Technical Specification Action Statement would require a plant evolution such as shutdown, but not always (Gallagher, July 1985). -
fecond, the current definition of OPERABILITY leads a licensee to declare a ,
r component inoperable when the component is only slightly degraded. In some j instances while a component is slightly degraded, it could'probably perform its function if it were needed;
- In each of the above situations, considerable difference exists in the safety
! significance of the "inoperability" problem as contrasted to a condition where the component is broken and can not function at all. In some cases.
i these problems could require shutdown to begin within one hour.
.i i
These probicms occur more than 100 times per year for U.S. reactors as a f whole. Most of the time, OPERABILITY is reestablished or emergency relief is I
obtained from the NRC, and the reactor is not shutdown. However, many times 4
the shutdown process is begun before relief is obtained. Any evolution such as reducing power or beginning shutdown involves some risk that an inadvertent trip could occur and a safety system will be challenged. Thereforb,an evolution which is not necessary from a safety viewpoint and which is caused j by these problems is itself an unnecessary risk.
4 i
- 5 2.-
l-l In addition to'the safety implications, th'ese inoperability problems require unnecessary expenditure of utility and NRC resources to resolve them. .TSIP has concluded that'some other method of dealing with these problems is 4
desirable. The method should balance the significance of the problem against the impact of the. transient required by plant shutdown.
i I 5 DEVELOPMENT AND EVALUATION OF ALTERNATIVE SOLUTIONS i
! After identifying the problems currently associated with Technical l Specifications, TSIP " brain storming sessions were used to identify and list potential alternative solutions to the problems. TSIP deliberately set out to explore a wide range of potential solutions.
l An examination of the list of alternatives showed that most overall solutions could be broken into components which addressed a particular problem, set of 4
problems, or a piece of a problem. For example, developing and implementing
[ criteria to determine what items should be in Technical Specifications l addresses the problem of lack of criteria but does not address the problems of i
- clarity and inadequate bases. However, criteria, in combination with a f
line-by-line review and revision of Technical Specifications, would address l'
both problems. The solution components identified by TSIP are listed under proposed solutions in the Evaluation Matrix (Table 5.1). ;
l After the alternatives were identified, a method of evaluating them was needed. TSIP identified three areas of consideration for evaluation of l potential solutions:
i l
l
! ^
I j . (1) Safety I (2) Resources
- (3) Administrative
)
t J
Using these areas of consideration, development of an evaluation tool i was begun. The Evaluation Matrix presented in this report is the final 1
i product of that development. Basically the matrix still uses the same three areas'of consideration, but a number of refinements and iterations were necessary to arrive at the final matrix. The Evaluation Matrix presented here addresses the potential solutions in these three' areas.
5.1 Safety Imoact 1
i i In the safety impact category, TSIP considered whether the proposed '
solution would improve or degrade plant safety. Changes which would achieve i
the following results were considered by TSIP to be safety improvements: j i
l Clearer language in the Technical Specifications.
P j Overall greater NRC attention or control (this is somewhat
- arbitrary but was included to reflect existing staff perceptions l and to recognize that NRC attention generally results in greater i
- overall attention).
j Focusing NRC or licensee attention or control on the most l important safety items.
4 i
1 1
L _ - - - - . - - - - _ - . ___ _ _ - . - - . - - .. - - _ _ - - __ . - - _ - .
54-i More complete technical Bases for Technical Specifications. !
Identification and correction of requirements for inappropriate
- . i
! tests or test methods or for testing more. frequently than is !
1 necessary. Safety can be degraded in several ways by testing: t t
- by making the equipment unavailable during the test, by wearing out-equipment, by inadvertently returning equipment to service in an inoperable condition, or by. causing challenges to safety systems (for example, an inadvertent trip due to operator error in
~
realigning system).
i
!' Identification and correction of action statements which may F
require licensees to take actions which are not appropriate from a 1
safety standpoint. In some cases, shutdown may not be the most j appropriate action because there is some risk of inadvertent challenges to safety systems even during an orderly shutdown.
7 Aids which contribute to a clearer understanding or usability of
}I the Technical Specifications, such as cross reference tables organized by system or mode.
4 Identification and correction of operability requirements and .
i .
action statements for inoperable equipment which tay be more l< P t
! stringent than necessary (For example, instructions that a system !
t
{ is inoperable if a surveillance requirement has been missed. Such an instruction may cause a plant to begin shutdown unnecessarily, again incurring some small risk of inadvertent safety system
]
challenges.) i l
en r, y-~ m._,.-.- .-%. , . .-_r -
g,m._-e,v_.,_.-,, .,,,wy,#,,,w.,_.--n m,v,,,r.,.-,- , . . , . , . , _ , w ,,.ww-.---am,,,,gw-y%.ww-,,,
l l
Any change which would achieve results opposite to this list was considered a reduction in safety by TSIP.
5.2 Resource Impact In the resource impact category, TSIP considered the costs in terms of money and personnel.both to licensees and the NRC to implement the proposed solution. The resource impact category was subdivided into short-and long-term impacts. For the purposes of these evaluations, TSIP graded the solutions based on the efforts that would be involved in the solution and TSIP's estimate of the magnitude of those efforts. These estimates were verified with quantitative cost or manpower estimates where the information was available.
Short-term costs were considered to be the costs to complete the implementation of the solution. Generally, the time frame for implementation of any viable solution would be no longer than 1 to 2 years.
i All short-term resource impacts were judged to be zero or negative because TSIP concluded that implementation of any change (solution) would require expenditure of additional resources before resource savings, resulting from the change, would be realized.
Efforts which could be required to implement solutions include:
1 1 -
56- '
I J
Rule changes.
1 t
i j Legislation. f l
f PRA analyses of A0Ts and STIs.
j Large scale revisions or reviews of Technical Specifications to
! - identify and correct linguistics, clarity, technical problems
! and/or to implement criteria defining scope.
I i
l Large scale generic PRA analyses and/or plant specific PRA j analyses.
i I
i Development of Technical Specification selection criteria.
l 1
Development of new controls for changes by licensees in procedures l
-i and FSAR (such as 10 CFR 50.59),
I i
Preparation of Technical Specification amendments, CRGR packages, :
i Generic Letters, NRR Offico Letters, Commis~sion Papers, Inspection
{ Manual revisions, Standard Review Plan revisions, NUREGs, ,
Standard Technical Specifications', and Commission Policy Statements. -
1 Allotment in NRC/NRR resources for a group dedicated to Technical j
i Specifications.
I I i
,-e--e-- -e- , e - nn ,x,.-,r,,, - - - ,,-c,,, ,-g., - , , , , -r-,,,,,, -,m-----_,-,--,---- --,v,, _w-+r.,---,-e- , - + -.rnv-m--,- -w,-,,-,w-r--.-----,- v---
. . . . .- .-. .. _ ~ .- - .- - _. . . - - - . _ _ _ .
i
~
4 Development of new prograwatic documents, such as an operational 1
surveillance progr'am.
J I
Large scale revisions to operating procedures, FSARs, or existing l programmatic documents such as the ODCM.
i Large scale revisions of Technical Specification Bases.
i l Complete rewrite of Technical Specifications, including s
j reformatting. ,
i i
i
" t j Licensee retraining i Long-term costs were considered to be those costs associated with whatever i
{ ongoing program was left in place after the implementation stage or short-t j term. Under long-term resources, TSIP considered who those costs would be (using TSIP's estimates, not quantitative values) m compared them to the
) costs TSIP would expect in the foreseeable future undet tha present Technical
} Specification scheme. t
! The following efforts currently consume-licensee and NRC resources and would I continue to do so under most solutions:
Technical Specification amendment applications, reviews, and 4
processing. Both the number of amendments and ease of~ handling
{ technically and administratively affect consumption of resource $.
j
)
i l Equipment testing and plant operations.
f f
o Processing, reporting and reviewing changes to FSAR, procedures, and programmatic documents.
The evaluation compared the current level with the expected level after a proposed solution was implemented. Those solutions that resulted in the lowest resources expended by licensees and the NRC received the highest positive scores.
5.3 Administrative Impact The administrative impact category is not as well defined as the safety and resource categories. In evaluating the administrative impact of the proposed solutions, TSIP considered the following questions:
Does the solution require a rule change or legislation? Both of these activities have large administrative burdens associated with them. They are time consuming and take a long time to complete, maybe more than the 1 to 2 years assumed in the short-term resource impact evaluation.
Does the solution reduce NRR's Operating Reactor Licensing Action Summary (ORLAS) backlog, result in fewer new amendments in the future, or make amendments easier to review / process?
Does the solution provide a clear understanding of the roles of Technical Specifications and other regulatory tools?
. . - . - = . - - - - . - . ~ . _ . .-. -- . - . . - . .
I Does the solution. provide a clear understanding of what items should be included in the Technical. Specifications?
)
- j. Does the solution complicate enforcement? ;
l i
I
- Does the solution provide a clear understanding of what NRC.
l lI group (s) has the authority and responsibility for Technical i
) Specifications?
f 1
4~
Will the. solution be acceptable to licensees, public interest groups, and the NRC Staff?
- Does the solution require development of new procedures for
! t j control of licensee changes to FSAR, procedures, or programmatic ;
! documents?
I Does the solution provide better understanding of the technical i i
l bases of the Technical Specification or provide technically l sounder Bases?
i i ,
I 5.4 Evaiuation Scoring ,
)
The matrix shows a score for each solution and impact. The scoring was done
-4 by relative judgments and a range of -3 to +3 was used. I i
I i !
E
.- . . . . . . - - . -_ .. .= -
a I
A +3 score means the solution provides the most improvement in comparison to 1 i the other possible solutions in the impact category under evaluation. Under safety impact, a +3 score means the largest-increase in plant safety over 1
the current level. In the resource impact, a +3 score means the greatest reduction in cost (dollars and manpower) from current levels. In the i
administrative impact, +3 score means the greatest reduction of administrative burden over the current level.
4 A 0 score means the solution causes no change in the impact category under evaluation. A -3 score means the solution causes the largest undesirable change in comparison to the other possible solutions in the impact category j under evaluation.
i :
i j More than one solution can'be scored with a -3 or +3 in a given impact category. ;
f This simply means that the impacts are of the same order of magnitude and that a finer scale would be needed to differentiate the impacts. These scores j were not arithmetically combined to produce a single, overall score for each i proposed solution. They were used to direct attention to the proposed solutions that presented the most positive impacts.
i i
i 1
! 6
)
i i
i 4
i
__ _ _ - ~ _ . . - .
- 1 A8t f 5.1 .
IVAttlAll0N MAIRIX FOR PROP 05tB 50lullf".510 ifCHNICAL SPICIFICATION PRoellMS Identified T. 5. Problem Proposed Solution Resource Impact Safety impact 5fi/rl Term long Tern .
A1.ainistrative Imp g
- 1. Riluctance to use other 2
'tio change, heavy use of T.S. to 0, 0, regulatory tools formalize conditions n uld continue 0, 0,
- No change by *No change by *No change by definition *No change by definition definition definition
- Number of I.S.
& amendments
' would centinue to increase
' Mechanisms such I as CNGR and Backiit Rule would act to control increase i
- Establish guidance on the roles of 0,+1, i.5. and other regulatory tools -1, +1,0, +1,0,
- Might help focus ' Staff effort to " Hight result attention or control *Might result in fewer prepare NURIG, in fewer new on most important ne w aneminents.
generic letter, amerwinents Pauviele a clearer safety items or Commission Policy Statement understanilinal of roles
- Rule change of regulatory tools probably not needed a
" Revise NRC controls associated o,,
4 +1 -1 teith other regulatory tools; controls 0,-1 3, Y!
- 0verall greater *MIghtrequirea could range from total NRC control NRC attention or *AddItionaleffort *Probably requires rule '
like 1.5. amendments to mainly change to rules required by change control such as 50.59 licensee controlled methods like controis ongoing *Might result in fewer 10 CIR 50.59. *Would require *Might result in development of 1.5. amendments fewer T.S.
- Requires develoceent new controls amendments of new procedures for
- Preparation of c harujes generic letter and *Would provide some CRGR package clearer understaruliruj
- Revision of of roles of regulatory inspection tools manuals
- lhis solution alone would tot be accepted '
easily t,y licensees 1 3 9
4
i l
identified T. 5. Problee Proposed Solution Resourca impact Safety impact Short Tere
- 1. tone Tere Administrative Japact Continued *Elle'inate T.S. Place conditions -1, -3, in other documents +3, -3,
- Less NRC control or Would require *No T. S.
attention rule change and amendments
- No 1.5. amendments
- Would really not legislation *May require
- Reduce ORtAS backlog address clarity and *Large revisions ' Requires rule changet technical probless additional effort and legislation unless specifically to procedures, and for new ongoing
- Unacceptable to NRC other documents change controls and intervenors reviewed in 'CRCR package.
conversion
- Complicates enforcese SRP revisions. *Provides clear under-inspection manual revisions standing of 1.5.
- May require *Might require new development of new change procedures for programmatic documents programmatic documents and new change grocedures Substantial licensee retraining
- Reduce T.S. to a minimum set of 0, 3, general requirements with conditions *Might reduce overall *MIghtrequire
+3 +2 handled in plant procedures, FSAR, NRC attention or rule change
'8Igreductionin *M$ghtrequirerule regulations,.and/or programmatic control T.S. amendments change documents *Large scale *May require
- Reduces usability ' Reduces ORLAS backlog problems of T.S. revisions to T.S. additional effort and other
- Reduces T.S. amendmenG to new ongoing *Provides clear under-documents such as change controls standing of what should
- Would not address procedures clarity and technical
- Preparation of be in T.S.
probless like CRGR package, $RP *$olution alone would inappropriate tests revisions, and not be easily accepted taless specifically in:pection manual by MRC staff (without reviewed in revisions additional controls)
, conversion to *May require *Might require new other documents development of change controls a
new programmatic
' documents and new $
a thange procedures
' Substantial licensee retraining k h e
4 e
_ - . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ m
identified T. 5. Problee Proposed Solution Resource tapact Safety Impact 5hort Tere
- 1. Continued tone Tern Administrative Impact
' Revise NRC framework for handlinq + 1, 0, 1.5. and amendments - NRR group given *Could focus attention
-1, +1, 0, + 1, 0, total authority and responsibility *NRR resources *!aplement on most important items for group
- Clearer lines of
- Probably additional order authority ard preparation of and control in responsibility T.S. amendment *No rule change various guidance grocess
- Night reduce documents Night reduce number of new number of new amendments 1.5. and
- 2. amendments Lack of criteria for 'No change, decisions made by inclusion of items in 0, 0, T.S. reviewers based on NRR Office *No change by 0, 0, Letter #34 - heavy reliance on *No change by *No change by definition definition 'No change by 1.5. to formalize conditions definition definittua
- Growth of T.S. .
would be controlled to some extent by CRGR and Backfit Rule
' Institute greater control of + 1, 0, 1.5. by organization - NRR group -1, + 1, 0, given total T.S. control or *Could focus attention 'NRR resources 'laplement *+1, 0, directives / guidance to NRC on most important for group . Clearer lines of managers. llees
- Preparation of additional order authority and guidance for and control in responsibility 1.5. amendment 'No rule change managers grocess
- Night reduce number Night reduce number of new T.S.
of new amendments. E u
and amendments '
8
Identified T. 5. Probles Proposed Solution Besource Japact Safety Impact 5hort Iere
- 2. Continued tone Tere Administrative Impact
" Eliminate T.S. Place conditions -1 -3 in other documents +3 -3,
- *tess NRC control or *Would require "No 1.5.
attention rule change and *No T.5. amendments
- Wuld really not legislation amendments
- Reduce ORLAS backlog address clarity and *Large scale
- hy require ' Requires rule change >
technical problems revisions to additional effort and legislation unless specifically procedures and for new ongoing
- Unacceptable to NRC reviewed in change controls other dociments and intervenors conversion 'CRCR package SRP
- Complicates enforceer revisions, *Provides clear under-Inspection manual standing of T.S.
revisions *Might require new
- May require change procedures for development of programmatic documents new programmatic documents and new change procedures
' Substantial licensee retraining
- Reduce T.5. to a minleus set of 0, general requirements with conditions 3, +3
- Might reduce overall *MIghtrequire +2 handled in plant procedures, FSAR *SIgreduction *Mlghtrequirerule regulations, and/or programmatic ISC attention or rule change dociments control *Large scale in I.S. amendments change
' Reduces usability *h y require ' Reduces ORLAS backlog problems of T.S. revisions to T.5.
and other additional effort for new ongoing
- Reduces i.5. amendment (transfers probles *Provides clear under-documents such as change controls standing of what should to some extent) grocedures
- Would not address Preparation of be in T.S.
clarity and CRGA package, " Solution alone would technical problems not be easily accepted like inappropriate 5RP revisions by NRC staff inspection manual tests unless revisions (without additional specifically *May require controls) reviewed in development of *Might require new ,
rn conversion to other dociments new programmatic change controls p documents and new
- change procedures .
- 5ubstantial licensee retraining 4 a
- e 4 .
E
.t: m t
c nge t S. ee o r at a aT p c
- . a p en h o f
et n da wal ab p a a rh r t uyc e l e l
b uc rf v p e
7 v Sl e5. :aeest r
ag bdl oa acl rmasu a c c
i t A u1 enderpso a L r a R lil rt eetC A r Oeet wscduinds R R t
s ,eb l
n n e nouo a h q clll ytN y Pa i
2 ci r mi edts se aaunrti in +,dswnoemynd usedvrs rwcncit eiai sse er m f yu t d aAf ndoi
+" e** o ea*e ruI* d ehfo*RBi D d mi*N cr a n A 1RPF mPnt t r d r n yy o ogs e f nl .is ae 5.a e,ml fi
. eeorgt1o r mb n S.sil Tt nn ssrl r nqnuaooetsee e ,eeeow ckaec netw.
t c g T
2+ cudm ritng e edrrna m m o t d sr a n ,d n yi nynpoee ee draae I
m p o t 1RmmdohMmoSx
+* a* af c* a t* e2 ht y e t t g
c r e pi n n l
ol a i t l u u l p f n n aw o r es l
pse ic i lf e ues s
e v aom u nnl R
e e
r ted ndn e 'A rAnr rR uPvn iRee t encemdr roog
,a o i
a a nt u ee eeg r iPq e u co q i t a qcrvs etidanan k o o T e f i ei i en e a e ci n c )
t reitr gnre rfdsso sr cato n a n pci e e r il ny sepa e s gb o , yn go t .ycuel e i c e d R pi n y h
s 3hh5f
- c* a i f ei S.
eot a n si rig s v a a cT*m pW x n o a L P u R n e h M r
s* eadb** gC i r c (
l d so sr n ot a l n n o t c otsh t
i s m gC c t oeiR :
a nmtlN e is p t n l i
m ta ot el nl a y l abr soti e t
e urrsv .
f ct oso
,onpo a 1FomPf S +" ci' o f
or e
th t ntnn t
eooe m n o d s'
- 4 ot es cnse r iasbs os ft a a
n uik o aors i p ei i
t rot r eri l
u tdrc i
ci S
o re t cc;s asi d pltl s e oPnie s l eb s o
p e may o v .ubl r De 5. coa orn .
P "1dpa s
e l
b o
r P
5 d
T e u
d e i n
i t f n i
o t C n
e d .
i 2 ll J !
identitled T. 5. Problee Proposed Solution Resource lapact Safety Impact _Short Iers
- 2. Continued long Ters Administrative lepact
- Develop criteria for content of T.S. O, -2, -1 +2,+1, Place dropouts into other documents; *Possible slight loss 'my re, +2,+3, criteria based on licensing analysis. of overall NRC control change (quire rule amendments
' Reduce T.5.
- Reduce 08tAS burden eventually)
- Focus attention and *5ignificant effort *m y require *Possible rule change control on most to deveitp criteria additional
- fewer new T.S.
important itees and split T.5. amendments effort for new *Provides clear revise other ongoing change documents understanding of what
- Preparation of controls items should fse in T.S.
guidance document, *5ome difficulty gettir CRCR package, NRC staff and interve inspection manual acceptance revision *Ny require developmen
- May require of control procedures developeent of new documents or change control grocedures
- 3. Some licensee Human factors & technical *No change retraining groblems. 0, + 1 inadequate Bases. t.hanges (am;endments) they feellet licensees propose 0,
- No Change by 0, 0,
' Clarity are necessary *No change by *No Change by definition definition *No change by definition
' Test methods and intervals (licensees might definition *0RLAS backlog would
- 0perability definition make improvements) *May result in probably increase C
ADT and inappropriate nore 1.5. amendments action statements
- Improve T.S. Bases Section + 1, *2, - 1, -2, +1, +2, s
' Provide more complete technical bases
- Effort to review *WI)) make 1.5.
+ 1, + 2,
- No rule change $'
- Help focus attention and revise bases amendments easier *Would help sake e
- No rule change to review on east leportant safety *Probably result enforcement less items complicated
- May identify in some reduction ' *mkes T.S. amendments inappropriate test of testing easier to review methods and intervals *Provides better under-standing, technically
- sounder bases v I e
9
- =
identified T. 5. Problee Proposed Solution Resource lepact .
Safety lapact Short Tere long Ters .
- 3. Continued Adelnistrative lepact
- Review T.S. for clarity problems +1, - 2, - 1, +1, +1, "Provides clearer
- Review and *leproved 1.5.
language in T.S. *No rule change
'. revision effort clarity will make
- No rule change plant operation a
- Licensee little easier retraining
- Reduce the need
! * *Possible CRGR for T.S. amendments package preparation
- Review 1.5. for technical problems *2, +3
' (testing operabilit -2, -3, *2, +2, +3, actions s,tatements) y A0i's, *Technlcalproblems ' Review and such as inappropriate
- Improved T.S., No rule change revision effort result in easier tests would be *CRGR package tpention and *Iewer 1.5. amendments identified and greparation less testing
- Reduction of ORLAS corrected No rule change backlog
- Large reduction " Improve enforcement
- titensee in need for i retraining *Provides better amendments understanding or technically sounder bases 4-* . I 01 9
k e
D e
Identiflod T. S. Problem Proposed Solution Resource Impact Safety lapact Short Ters
- 3. Continued tonq Iern Administrative Impact
- Rewrite T.S. Including new +3, formating, clarity and technical . -3, +3, review
- Improved T.S. clarity
- Hay require rule +3,
'Hore complete change ' Result in fewer
- Hay require rule chan technical bases
- largest T.S. T.S. amendments
- Reduce ORLAS backlog
- Identification and
- Easter plant revision effort operations and
- Result in fewer T.S.
correction of technical
- Preparation of less testing amendments problems such as " facilitate handling c inappropriate tests CRGR package, SIS,
- facilitate amendments genericletters handling of T.S.
ticensee amendments 'Provides better under-retraining standing of T.S. and better technical bases.
- Makes enforcement less complicated Provide aids such as cross + 1, 0, reference tables for systems - 1, 0, +1 0, and modes *Some improvement in *Relatively small *[a, ster plant +1, 0, clarity, help ensure *No rule change items not missed effort to implement operations
- No rule change
- Improve clarsty Use PRA to evaluate A0T's and + 1, + 2, Sil's (allowed out-of-service -2, -3, +2,
'Hore complete + 1, + 2, times and surveillance test
- Performance of
- Reduce T.S.
intervals) technical basis for PRA analyses to *No rule change T.S. amendments evaluate A0l's and *[asier plant
- Reduce 1.5. amendments
- Correction of overly SII's and NRC operations
- No set PRA acceptance restrictive A0i's and review *tess frequent criteria Sil's *Hore complete safety testing bases for T.S.
s 4 I e
e e
O
o APPENDIX A BACKGROUND ON PURPOSE OF TECHNICAL SPECIFICATIONS To provide adequate protection to health Atomic Energy Act 1954
& safety of the public - common defense and security.
To reasonably protect the integrity of the 10 CFR 50.36 physical barriers (fuel, RCS, containment)
(safety limits, limiting safety system settings).
To~ prevent alteration of design features that 10 CFR 50.36 would have a significant effect on safety (design features).
To assure the necessary quality of systems and 10 CFR 50.36 components is maintained, and that the facility operation will be within the safety limits (surveillance requirements).
Define and preserve those underlying Proposed Rules, assumptions that are expected to, or could, Fed. Reg. 03/30/82 vary with time or circumstances, through the life of the plant, and thus to preserve the validity of the safety analysis.
i To validate the assumptions of the safety Corcoran, Combustion analysis that are not validated by other Engineering, " Verifying means. When these assumptions are the Adequacy of Technical validated, the risk is controlled to that Specifications" level found acceptable in the licensing process.
Are to be reserved for those matters as to ALAB-531 !
which the imposition of rigid conditions or limitations upon reactor operation is deemed necessary to obviate the possibility of an t event giving rise to an immediate threat to the public heelth & safety.
Technical Specifications should.be limited to Marvin Mann those aspects of the reactor system which bear Reg. Review Panel a direct relation to public safety, rather
! than a detailed description of all components of the reactors.
t The Technical Specifications will reflect in 1962 Rule l
such form limits in design and procedures approved by the Commission. They will represent,
- in essence, those parameters which define the boundaries of licensed activity which the Commission has evaluated and approved from a 3
safety standpoint.
J
, _ - . . _ , - . . , - - , - - - - - , - , - _ - . _ , . , . _ - - -, _ , , , - . , , _ . _ . , _-,_,,,c,,. - - - . . - . - - , , . , _ . , . _ - . . -
i 6 REFERENCES 4
I Law Atomic Energy Act of 1954, as. amended. 42 USC 2011, et. seq. at 2232.
Regulations
, Title 10 Code of Federal Regulations, Part 2, Appendix C, General Policy and _
Procedures for NRC Enforcement Actions.
Title 10 Code of Federal Regulations, Part 50, Section 50.36, Technical Specifications.
l Title 10 Code of Federal Regulations, Part 50, Section 53.59, Changes, Tests, and Experiments.
Proposed Regulations i 47 FR 61, March 30, 1982, " Technical Specifications for Nuclear Power Reactors."
i Board Decisions I
4 i
In the Matter of Portland General Electric Company (Trojan Nuclear Plant).
- f. ALAB-531, 9 NRC 263, (1979) et. seq. 273.
l j '
i NUREG Reports NUREG-1024, " Technical Specifications - Improving the Safety Impact,"
l November 1983.
1
- Reports 1
Azarm, M. A. , J. L. Boccio, R. E. Hall, S. Karimlan, " Operational Safety l
Reliability Research Project Plan", Brookhaven National Laboratory, i
June 1985.
I Corcoran, W. R. , S~. A. Webster, J. C. Braun and D. R. Earles, " Verifying the Adequacy of Technical Specifications," Transactions of the American Nuclear l Society 1983 Winter Meeting, 43.555-6, November 14-18, 1982.
Gallagher, D, and B. Atefi, " Impact of Technical Specifications on
- Operating Reactors During 1984" SAIC-85/1777, July 31, 1985.
1
}
i Jansen, R. L. , L. M. Lijewski, and R. J. Masarik, " Evaluation of Surveillance Frequencies and Out of Service Times for the Reactor Protection Instrumentation System," WCAP-10271, January 1983. -
i l Phung, D. L. , " Impact of New Reactor Safety Information on Technical i Specifications," Professional Analysis, Inc., August 15, 1985.
1 1
(
D O
Samanta, P.'K., S. M. Wong, and J. Carbonaro, " Evaluation of Risk Impacts of ANO-1 Technical Specification Requirements," Brookhaven National Laboratory, August 23, 1985.
1 Sullivan, W. P. , " Technical Specification Improvement Analyses for BWR Reactor Protection System," NEDC-30951, May 1985 (Proprietary Information.
Not publicly available.)
1 1
Letters i
Eisenhut, D. , NRC, to All Licensees,
Subject:
" Clarification of the Term
- Operable," April 10, 1980.
Kripps, L. J. , Energy Inc. , to T. Collins, NRC,
Subject:
" Review of
- Standard Technical Specification Containment System Bases,"
q September 17, 1985.
l Thomas, C., NRC, to J. Sheppard, Westinghouse Owners Group,.
Subject:
- " Safety Evaluation Report on WCAP-10271," February 21, 1985.
1 Memorandums i
Beckham, D. H. , NRC, to D. G. Eisenhut, NRC,
Subject:
" Program Plan for the Technical Specification Improvement Project," March 19, 1985.
Beckham, D. H., NRC, to D. G. Eisenhut, NRC,
Subject:
" Technical
! Specification Improvement Project - Interim Report," August 29, 1985.
. . , , _ _ . _ _ . _ . - . . _ . , _ . _ _ . . . , _ ...._._,,_.,___.,,______m.,__....._,____.m,,,._,,_,.-...__ c~..,_,_
Denton, H. R., NRC to All NRR Employees,
Subject:
"NRR Office Letter 34 -
Utility Commitments," July 1981.
Denton, H. R. ,~ NRC, to All NRR Employees,
Subject:
"NRR Office Letter 40 -
Management of Proposed Generic Issues," March 1983.
Denton, H. R. , NRC, to T. P. Speis, NRC,
Subject:
" Formation of a Technical Specification Improvement Project Group," December 31, 1985.
Dircks, W. J. , NRC, to H. R. Denton, NRC,
Subject:
" Report by the Task Group to Study the Design of Surveillance Testing in Technical Specifications,.NUREG-1024," November 14, 1983.
Rowsome, F. , NRC to W. Minners and S. Bryan, NRC,'
Subject:
" Candidate Generic Safety Issue: Allowable Outage Times for Diverse, Simultaneous Equipment Outages," May 9, 1985.
O
- .y .
E 9
R
- !iii N
INCOMING AND SIGNATURE TAB Use this side of the sheet to precede the incoming material when assembling correspondence.
(USE REVERSE SIDE FOR SIGNATURE TAB)
I
.J r
s 4
l
> e Ata m63 Ind zstrill Fa rum, Inc.
7101 Wisconsin Avenue Bethesda MD 20914-4805 Teleprone (30t1654 9260 TWA 7108249602 ATOYlC FOR DC
_J
~
October 8, 1985 -
Mr. Harold R. Denton Director, Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington,~ D.C. 20555
Dear Mr. Denton:
Enclosed is a copy of the report entitled, " Technical Specifica-tions Improvements", prepared by the AIF Subcommittee on Technical Specification Improvements of the Committee on Reactor Licensing and Safety. The report represents the results of over six months of intense effort by the Subcommittee, its four working groups and many discussions with your staff of the Technical Specification Improvement Project (TSIP). The Subcommittee consists of the Chairman of the four vendor Owners Groups Technical Specification Subcommittees, the Chairman of the Nuclear Power Plant Standards Working Group, ANS-58.4, as well as representatives from EPRI, individual utilities, AE's and consultant firms.
As Mr. Alan Passwater, Chairman of the Subcommittee, highlighted 4 during the October 1, 1985 meeting with you and your staff, Sec-tion II of our report provides the recommendations and conclusions of the Subcommittee. These recommendations include improvements that can be made now within the present regulatory framework. We strongly encourage the NRC to proceed immediately to work with the Owners Groups, individual utilities and the Subcommittee in imple-menting these improvements. Specifically, NRC endorsement of the recommended criteria and improvements should be obtained as soon as possible.
In order to gain the necessary improvements discussed both in our report and the TSIP's report, the necessary staffing resources and management attention must be dedicated to this very important issue. The organizational responsibilities for implementing the recommendations and the individuals responsible for reviewing and approving the requested changes in technical specifications, either generic or plant specific, should be delineated as soon as possible. These individuals should work closely with the Owners Groups, individual utilities, and the Subcommittee to develop schedules for timely implementation of the recommendations and reviews of existing and future requests.
I i
Mr Harold R.,Denton October 8, 1985 -
- 2 l ! I-a -
l 1
One of the major differences in our recommendations and those of l the TSIP is the need for rulemaking to gain full benefit of the !
improvements that can and should be made. Although, the TSIP did not recommend changes in the regulations, we feel that there are i changes that should be made in them to obtain full benefit of improvements in technical specifications. I must emphasis however that improvements that can be made now should not be delayed until any changes to the regulations are made. These two activities can and should go forward in parallel and on separate schedules.
The Subcommittee and its Working Groups have found the working.
relationship with the TSIP staff in addressing this very important topic extremely useful in coming to a mutual understanding of the problems and suggested recommendations. We encourage continua-
- tion.of this approach during implementation of the recommendations and stand ready to continue discussions during this very important phase. The Subcommittee has spent considerable time in developing
, the criteria and the recommendations and looks forward to pursuing
- the improvements recommended. As soon as you and your staff have j had an opportunity to review our report, I would suggest a meeting be scheduled in early_Nov_ ember _to_ discuss its content and how best 3
to implement the recommendatfons. _
I Sincerely,
[b hw Murray R. Edelman
, Chairman, Committee on Reactor i Licensing and Safety MRE:tkr
! Enclosure f
i I
I l
l i
.- =. . -- - . .. _ . . - _ _ - - . . . . _ - - . - . . - . _..- _
i.
l I
T i
2 l l i . :
, Technical Specifications i Improvements
!. ~
t i by AIF Subconnittee on !
Technical Specification Improvements of the
~
- Committee on Reactor Licensing and Safety J
1 1
4 :
w 1 ,
i i
4 October 1, 1985 I
t, 1
I 6 l I i
i l
i l l
t
(
4 4
l 4
Table of Contents
- 1. Introduction .
A. Report Purpose B. Report Content C. The Evolutionary Background of Technical Specifications D. New Initiatives for Technical Specification Improvements E.
AIF Subconunittee on Technical Specification Improvements
- F. Probabilistic Methodology II.
. Conclusions and Recommendations III.
Improvements in Technical Specifications To Alleviate Current Problems A. Promoting Safe Plant Operations B.
Facilitating Licensee Compliance with NRC Requirements C.
D.
Minimizing Manpower and Paperwork Burdens for the NRC and Licensees Increasing Plant Performance / Availability IV. A New System of Technical Specifications A. The General Philosophy of Technical Specifications B. Selection Criteria for the Present Technical Specifications C.
Selection (LCOs)
Criteria for the Present Limiting Conditions of Operations D.
Disposition Specifications of Requirements not Appropriately Contained in Technical V. Implementing the New System A. Applicability of the New System B. Changes to the Regulations C. Short Term Improvements D. Conversion of Technical Specifications E. Format and Style of New Technical Specifications Appendix A:
Regulatory History and Current Regulatory Requirements and Guidance Appendix B:
The Paperwork Burden of the Sholly Regulations Appendix C:
Public Consnents Technical Regarding the 1982 NRC Proposed Rulemaking on Specifications Appendix D: Establishing Critieria Appendix E:
Examples of Technical Specifications Under the New System Appendix F:
Disposition of Requirements not Appropriately Contained in Technical Specifications Appendix G:
Discussion on Short Term Solutions not Requiring Rulenaking Appendix H:
Suggested Changes to the Regulations Appendix I: Probabilistic Methodology I
Appendix J: Abbreviations and Acronyms
- I. INTRODUCTION 1
It is generally agreed that the technical specifications in effect at ~most nuclear plants today are in need of review and simplification. The importance of the key information contained in them is diluted by the large volume of information. Some
- of the requirements are actually adverse to safety, others are simply ambiguous. A program has been undertaken by both the industry and NRC to improve technical
~
specifications; to separate the most important from the less important, to clarify
~
! the content and to improve overall plant safety.
Report Purpose A.
This report articulates the industry's proposals for technical specification improvements and presents a regulatory basis for the proposed technical l specification reforms. It serves as a communication vehicle both among nuclear 1
industry personnel and between the industry and the NRC. Full implementation of the technical specification improvement program will require substantial resources from both the industry and the regulatory agency. This doctanent has been prepared to assure efficient utilization of the resources required during the implementation phase. This can be accomplished by having the NRC and industry reach agreement on *
- as many substantive matters as possible prior to the actual implementation and by i solving generic issues on as broad a scale as possible.
- i e B. Report Content i <
Section I provides background information.Section II presents conclusions and recomendations derived from discussions presented in the balance of the report.
, Section III describes the expected benefits from the full . implementation of l improved technical specifications. A detailed description of this new system is contained in Section IV and its implementation is described in Section V.
Appendices provide details and related infomation.
C. The Evolutionary Background of Technica1' Specifications The following information sumarizes the evolutionary history of regulatory actions related to technical specifications. Appendix A provides additional details.
l Section 182a of the Atomic Energy Act of 1954, 42 U. S. C. 2232(a) provides in part that:
l "In connection with applications for licenses to operate production or i utilization f acilities, the applicant shall state such technical J specifications, including information of the amount, kind, and source of special nuclear material required, the place of the use, the specific
- characteristics of the f acility and such other information as the
] Comission may, by rule or regulation, deem necessary in order to enable !
it to find that the utilization or production of special nuclear material '
l will be in accord with the comon defense and security and will provide l adequate protection to the health and safety of the public. Such technical specifications shall be a part of any license issues."
1
- 2. .
This statutory directive has been implemented in the Comission's regulations,10 CFR Part 50, Section 50.36, entitled " Technical Specifications." Before 1968, Section 50.36 required technical specifications to include "those significant design features, operating procedures, and operating limitations which were l considered important in providing reasonable assurance that the facility (would) be constructed and operated without undue hazard to public health and safety."
In December,1963, the Atomic Energy Comission (AEC), predecessor of the NRC, amended its regulations in Sections 50.36 and 50.59.(33 FR 18612). Section 50.36
- was amended to include a more precise definition of those categories of technical specifications that must be included in an application for an operating license.
The amended regulation narrowed the scope of the material contained in technical specifications by defining five specific categories of technical specifications.
The five categories defined for nuclear reactors are: (1) Safety limits and limiting safety system settings, (2) limiting conditions for operation, (3) surveillance requirements, (4) design features, and (5) administrative controls.
(47 FR 13370) ,
On July 8,1980, the Comission published an Advance Notice of Proposed Rulemaking (ANPR) requesting coments on the desirability of changing its regulations on technical specifications to: (1) establish a standard for deciding which items dcrived frcyn the safety analysis report must be incorporated into the technical I specifications for a f acility; (2) modify the definitions of categories of technical specifications to focus more directly on the aspects of reactor operation that are important to the protection of the health and safety of the public; (3) defina a new category of requirements that would be of lesser importance; and (4) establish a mechanism for the licensee to make changes without prior NRC approval.
On March 30, 1982, af ter considering the public coments made in response to its prsvious ANPR, the NRC published a proposed rule for coment. This is the so-called " George" rule to split the technical specifications, relocating some
! material to a document called supplementary specifications. In the supplementary infonnation, the Comission explained that a recent legal case "... highlighted the need to establish specific criteria in the regulations for deciding which items derived from the safety analysis report must be included in the technical specifications incorporated in the license for a f acility."
This rulemaking for technical specification reforms was never completed. The Comission suspended efforts on the rule around mid-1983 in order to concentrate staff resources on the issuance of near term operating licenses. Public coments regarding the 1982 proposed rule on technical specifications is provided in Appendix C.
D. New Initiatives for Technical Specification Improvements Beginning in August,1981, with the publication of NUREG-0839 (Results of a Senior Management Survey), the staff acknowledged the potential for safety reductions due to poorly conceived technical specification requirements including test frequencies i and scope. More recently specific concerns were raised in regard to reactor trip system test frequency and diesel generator test frequency and scope. These concerns were documented in associated Generic Letters. In August, 1983, the-Deputy Executive Director for Regional Operations and Generic Requirements directed i
i *
.. , _ __.. .. _ _ .-~ _ _ _ , , . _ . - , -
the establishment of a Task Group to review technical specification requirements.
This effort was documented in NUREG-1024, " Technical Specifications -- Enhancing the Safety Impact", November 1983.
The recomendations of this Task Group, along with other motivating f actors, led to the establishment of the NRC's Technical Specification Improvement Project (TSIP) in December,1984 The objective of the TSIP was to reconsider the entire area of technical specifications, including philosophy, scope, content, depth, and the process by which they are implemented and enforced.
To interf ace with the TSIP, the industry fomed a Subcomittee under the auspices Gf the Atomic Industrial Forum (AIF). While prior technical specification improvements had been on an individual utility or Owners Group basis, the AIF Subcomittee serves as a focal point for individual utility optimization efforts as well as Owners Groups activities on generic issues associated with technical specification improvements.
In developing an overall plan to address technical specification improvements, the Subccrrnittee has addressed the key elements needed for a successful program. This includes developing criteria, the administrative controls needed, proposed regulatory changes and near-tem solutions. The overall program plan is shown on
- Figure 1.
E. AIF Subcomittee on Technical Specifjcations Improvements On March 1,1935, representatives of industry met with the NRC to present the industry's proposal for improving technical specifications. This proposal included the development of criteria for determining the content of optimized technical specifications. The criteria will be used to split current technical specifications, either custom or standard, into optimized technical specifications. The existing technical specification requirements that did not satisfy the criteria would go into supplemental specifications or an updated FSAR.
Additionally, the industry comitted to fom an AIF Technical Specifications Subcommittee to coordinate industry plans and policies and interact with the NRC.
The AIF Subcorrnittee on Technical Specification Improvements was formed mid March, 1985 and reports to the AIF Comittee on Reactor Licensing and Safety. The four light water reactor owners groups (LWROGs), EPRI and ANSI as well as individual utilities, vendor and AE fims have representation on the Subcomittee. The Subcommittce established four working groups to develop recommendations. The b.alance of this report is the result of the efforts of the following working groups:
Criteria Development; Suggested Regulatory Changes and Administrative Process for Converting to the- Revised Approach; Administrative Process for Controlling and Maintaining the Revised Document; and Research on Technical Specifications; including Probabilistic Methodology and Criteria Applications.
~'
Figura 1 .
Pl.AN TO REFORM NRC REO.UIREMENTS REI.ATED TO THE TECHNICAL SPECIFICATIONS OLD SYSTEM NEW SYSTEM , ,
I I l
l 8
! I I I I
- ACO *3nemownon.cssentpAt. .
t I g g e AcornoMAL
- 2WMMLANCE y,"*,'y,^*
^
- g g ,,gg,",gy
,c 9 l l INTER 188ts Y -ntNOWDt309n n9 FORMA 9 son4 I I
. 'ILisTs, ese) g l
\
, g e oprnworArronet I :nemovepnoenannararse
- ALLO 998RLE OUTAGE eq DETAets I I TMES n f
,t _ -, .
/
SNORT-TERM e e e ConsotmArtismRewy he09ROWNENTS g g LIMIT 9Me CONOITIOMs
/ FOR OPERATION i
e i I' s I l
I
/ e l
1 SFWEOP AsegngtsFnAFFWE con 7Wot3 a l
' FOn NEnv system
} ,
- newst n90tKAfonY t i
\
REQUIREMENTS .
) g -
NRC ipCENSEE
\ APPRO908L IMPLEMENTAT90M
F. Probabilistic Methodology The Subcomittee is gratified to see that the NRC has, on a selective basis, considered the use of probabilistic analyses in justifying changes in the existing technical specifications. The industry encourages this work and plans to continue to interact with the NRC in the further development and application of this C;thodology.
~ Criteria for defining the scope and content of technical specifications.could have been developed through the use of probabilistic methods. However, it is not clear that PRA based criteria would provide a demonstrable incremental benefit in technical specification improvement. Properly defined deterministically developed criteria provide greater benefit. Because of this f act, and in recognition of the netd to maintain consistency with the current regulatory basis for nuclear plants, the Subcomittee has proposed a set of deterministic criteria to define the content of improved technical specifications. The Subcommittee supports continued investigation into the application of probabilistic methods and associated acceptance criteria for improving technical specifications, within the context of a .
deterministically-developed set of criteria for their content, such as those proposed in this doctsnent. Imediate areas for these applications are the assessment of Surveillance Test Intervals (STI) and equipment Allowable Outage Times (A0Ts), now found in technical specifications.
The Subcomittee believes that significant improvements in technical specification requirements for spe:ific systems and ccrnponents can be achieved now and 'it ,
recocynends that the NRC be prepared to accept, review, and approve requests for such changes (e.g., for STIs and A0Ts) which are based on the application of probabilistic methods, as well as for those requested improvements which are based on the application of more deteministic criteria for technical specification
- c ntent. The parallel application of deteministic and probabilistic methods for -
improving technical specifications is feasible and entirely appropriate. These cethods, when used in parallel, should be consistently applied by industry in i
substantiating future requests for technical specification changes, and by the NRC i
in evaluating these requests.
The industry and NRC representatives at an August,1985 meeting agreed that a continuing dialogue on the development of methods, procedures and criteria for application of probabilistic methods to technical specification improvement should be maintained. The industry representatives requested an opportunity to review the pr: ducts of the PETS program as they become available. In return, industry representatives, through the AIF Subcommittee on Technical Specification Improvements, agreed to provide coments on these products and other issues to the responsible organizations within NRC.
After extensive discussion within the Subcommittee on the uses and benefits of probabilistic methods in obtaining technical specification improvements, it was concluded that the establishment of the proper role of such methods in the process will be greatly f acilitated by their application now, where possible and clearly benefici al . The Subcomittee recomends that both the NRC and industry continue expanding the applications of these methods for technical specification improvement. Additional discussion on this topic is found in Appendix I.
II. CONCLUSIONS AND RECOMMENDATIONS The purpose of this section is to articulate the industry's conclusions and recomendations based on the material discussed in the balance of the report.
1 CONCLUSIONS l
- 1. The present scope and content of technical specifications do not necessarily enhance safe f acililty operations. Some requirements are confusing and present an element of unnecessary frustration to both the f acility operator and the regulator. There is no clear understanding of the purpose of the docment, what should be included and why, resulting in at times unnecessary exchanges between the licensee and the NRC with questionable benefit to public health and safety. There is a definite need to seek improvements in technical specifications, regarding both philosophy and content.
- 2. Substantial improvements can be made in technical specifications in the imediate future. These improvements are consistent with regulatory requirements that presently exist. Such changes include removing information which duplicates that presently found in other documents and in other regulations, providing clarification of terms used, and reconsidering the -
technical basis for certain requirements. These types of improvements can and should proceed irrediately.
- 3. Explicit criteria are needed to detemine which structures, systems, components and process variables should be included in the technical specifications.
These criteria will assist the staff and licensee in determining what elements of the f acility have a unique level of importance that warrants their inclusion in this document. For long term stability, these criteria should be explicitly written into the regulations,specifically 10CFR50.36.
- 4. Pilot studies conducted by the Owners Groups have successfully demonstrated use of the recomended criteria. (See Appendix D for detailed discussion)
- 5. There is no need to establish a separate document entitled " supplemental specifications" as suggested in the proposed rule (47 FR13369 March 30,1982).
Existing documents and associated administrative controls can be used for those topics no longer required to be retained in the technical ' specifications.
which must be retained in the interest of safety.
- 6. There are technical improvements that can be made in regard to such matters as allowed outage times and surveillance test intervals using probabilistic methodologies and operating experience, which provide overall improvements in safety and operability.
- 7. There is no clear benefit to public health and safety in requiring " cumulative outage times" in the revised technical specifications. The additional administrative burden and potential stress of returning equipment to operation too quickly in order to minimize expenditure of outage time allowances may compromise quality repairs with questionable benefit to the overall safety of the facility.
RECOMMENDATIONS
- 1. Criteri a A. The NRC should endorse the AIF technical specification criteria for application on a voluntary basis by utilities to existing and future technical specifications.
B. Following NRC endorsement of the criteria, each Owners Group should apply the criteria to their respective Standard Technical Specifications. These revised documents, with technical justification, should then be submitted to the NRC for review and approval for use as generic guidelines.
C. Pending staff originated technical specificat' ion additions should be reviewed against the criteria and those not meeting the criteria should be placed on hold pending resolution of NRC approval and codification of the criteri a. ,
D. The NRC should initiate rulemaking to codify the criteria for use on a voluntary basis in place of the current requirements of 10CFR50.35.
E. The industry and NRC should participate in preparation of a revision to ANSI /ANS 58.4 (Criteria for Technical Specifications for Nuclear Power Stations) to p-ovide a "w-iters' guide" to utilities on preparation of technical spe:ifications using the criteria. The NRC should consider endorsing this revised standard for use on a voluntary basis by issuance of a Reg Guide.
- 2. Process The basis for adequate control over items which are excluded from the new technical specifications f alls into three areas: ,
- 1. 10CFR50.59
- 2. Licensee Ccmitments
- 3. Federal Regulations Based on these controls, and the link established between the FSAR and plant procedures, the NRC should concur with the process as depicted in Appendix F, Figure F-1.
- 3. Short Tem Imorovements Appendix G to this report provides recomended improvements to several specifications which should be reviewed by the NRC. It is requested that the NRC accept and utilize these documents as the basis for generic guidance which would f acilitate plant-specific proposed license amendnents. Further improvements are under develogynent by the Owners Groups and will be submitted to the NRC for similar consideration. Upon NRC issuance of generic guidance on these issues, each participating utility should consider promptly responding with a proposed amendnent incorporating the changes.
Sufficient resources to review and approve requests for these technical specification improvements should be provided by the NRC. A single central organization within the NRC should be responsible for ensuring consistent application of these improvements.
1
- 4. Other Rule Changes In addition to the codification of the criteria outlined in recocnendation 1.D above, several confonning changes should be processed as described in Appendix H. These include, but are not limited to, such items as 10 CFR 50.36a (RETS),
10CFR50, Appendix J, etc.
- 5. Other Longer-Tem Changes In Recomendation 3 above, changes which are relatively straightforward and which do not involve a change to the rules are proposed. There are other changes which do not require rulemaking and which may be somewhat more complex, that should also be pursued. An example of such an item is improvements in the bases.
- 6. Probabilistic Methodology A. Sufficient resources to review and approve probabilistic requests for technical specification changes should be provided by the NRC. A single central organization within the NRC should be responsible for ensuring .
consistent application of these probabilistic methods.
B. The present dialogue between the NRC and the industry on the application of probabilistic methods to technical specification evaluations and improvements should continue. The opportunity for the broadest possible technical review of the PETS program products, within the NRC and the industry, should be provided.
C. Future applications of probabilistic methods beyond the evaluation of changes to surveillance requirements and allowable equignent out-of-service times should continue to be pursued.
D. The industry and the NRC should strive to achieve agreement on appropriate probabilistic methods (either generic or plant specific) and related acceptance criteria which can be used for review of technical specification changes.
E. Acceptable means should be devoinned tn ann 1v tha racnits nf a generic probabilistic analysis directly to individual plant requests for changes.
These means should not hecessarily require the existence of a plant specific PRA for each individual applicant plant.
~
III. IMPROVEMENTS IN TECHNICAL SPECIFICATIONS TO ALLEVIATE CURRENT PROBLEMS
- The overall improvements in technical specifications described herein provide i
a logical way to alleviate current technical specification problems and bring these individual efforts to fruition. It is expected that the program will have multiple benefits as described below.
- There are four anticipated benefits that can be realized from full
. implementation of the proposed technical specification improvement program:
~These are:
o Promoting safe plant operations; o Facilitating licensee compliance with NRC requirements; and o Minimizing manpower and paperwork burdens for the NRC and licensees, o Increasing plant perfomance/ availability The following sections discuss each of these benefits separately.
A. Prmotino Safe Plant Operations
- Contrary to the originally intended function of technical specifications, evolutionary chances over the past 15 years have resulted in the introd -ti of many requirements which actually may be adverse to safety. Concern 3bouin this possibility has been raised by both industry and NRC representatives. An NRC Task Group on Technical Specifications established in August,1983, confimed that there were several areas in which needed technical specification changes could probably result in direct safety improvements.
For example, the Task Group indicated that requiring surveillance testing of an operable train while equipment in the other train of a system is declared inoperable may "actually degrade the needed system and increase public risk."
The Task Group effort which was quite comprehensive was documented in NUREG-1024. This doceent is considered to be a good assessment of ways in -
3 which present technical specifications could adversely affect safety. One of the key objectives of this program is to correct these problems, thus resulting in improved safety. '
- A second significant way in which safety can be improved is through technical specification simplification.~ Much of the infomation presently contained in i technical specifications duplicates that found in other licensee documents and MRC regulations. A typical volme of technical specifications for a modern day nuclear plant is several inches thick. The table of contents alone may require as many as 25 pages. It is generally agreed that one of the key i values of technical specifications is to comunicate to operations personnel ;
cxactly what the important limits and limiting conditions are for the plant.
Technical specifications currently do not effectively prioritize or highlight the most important information. This dilutes the significance of the information which is of most imediate importance to assuring safety. Making the technical specifications easier for operations personnel to use is another way to promote safe plant operation.
! The impact of the substantial growth in the size of technical specifications was noted by the NRC in its proposed rule on technical specifications (47FR13369,3/30/82). It was noted in the preamble that, "The Comission is ccncerned that the increased volume of technical specifications lessens the likelihood that licensees will focuse attention on matters of more imediate importance to safe operation of the f acility."
p--. - - , - - . - - - - - . - , , - - --r- -,,.m-.--- ,,-w-- -
r a ,- - - , , 1, .w--, , . - - . . , - -
l j
j B. Facilita' ting Licensee Compliance with NRC Requirements Any effort that focuses the scope and clarifies the content of technical specifications will increase the attention on safety-significant areas. Relocating '
requirements in discrete documents (e.g. In-service Inspections-(ISI) plan) will avoid possible conflicts (real and imaginary) and will f acilitate understanding and tracking. The current situation with regulations referring to technical-specifications and technical specifications providing exemption to regulations as well as supplementing regulations (all true for 10CFR50, Appendix J'itself) is confusing and inappropriate.
The NRC's Office of Inspection and Enforcement had discussions with several licensees during April and May,19S1 to, "directly ascertain the perspective of licensees on the depth and scope of this negative safety impact problem." The results of this effort were discussed in NUREG-0839, "A Survey by Senior NRC Management to Obtain Viewpoints on the Safety Impact of Regulatory Activities from Representative Utilities Operating and Constructing Nuclear Power Plants," issued ir August, 1981. These discussions involved shift personnel, plant engineers, lower level plant supervisors, senior plant managers, as well as senior corporate canagement. In discussing the technical specifications, it was noted on page 12 of .
NUREG-0839 that, "According to one group of managers, NRC has ' lost sight of the forest because of the trees' in imposing new technical specifications. The required amount of detail has approached the unreasonable level. In a specific administrative example, it was stated that the format of the technical specifications changed six times in the last two years." 'On page 15 of the same document it was noted that, "The technical specifications, including reference requirements, continue to grow. Technical specifications are looked upon more as a trap than as safety requirements." There were many additional concerns expressed during these meetings.
The current technical specifications of ten have unclear or missing technical bases. This lack of bases promotes legalistic compliance in lieu of clearly understanding and meeting the technical concern. Therefore, bases improvements will also improva compliance and will help assess the significance of non-compli ance. An example of this is the ongoing dialogue between licensees and the NRC on Appendix J requirements.
C. Minimizing Manpower and Paperwork Burdens for the NRC and Licensees The process of amending many items contained in current technical specifications
~
now imposes an unnecessary and costly burden on the NRC and. licensees to meet administrative requirements which do not seem to serve the needs of the public or the nuclear industry.
Under the current technical specification amendment system implemented in 1933, the Commission requires an applicant requesting an amendment to its operating license (technical specifications) to provide its appraisal on the issue addressing whether or not the amendment involves a significant hazards consideration using the standards in 10CFR50.92, and, if the amendment involves the emergency or exigency provisions, to address the features upon which the Consnission must make its :
findings. The licensee also is required to notify the State authorities of the l l amendment request and the results of its significant hazards consideration review. l
The 1983 "Sholly" regulations, implemented by.48FR14873, has greatly increased the amount of administrative effort involved for both the utility and the NRC in issuing license amendments and introduced complications that can significantly expand the NRC effort. These complications include such things as: internal controversy between technical staff and legal staff which can necessitate extensive rewrites and meetings, disagreements between NRC staff and licensees, extensive coments on the Federal Register notices, and opposition resulting in the need to
. hold a public hearing.
A Congressional review of the implication of "Sholly" concluded in part that the public does not appear to benefit from these hearing opportunities for many items now included in the technical specifications. Many licensing mendments are sought to correct typographical errors, change corporate structure, or update equipment lists which are currently contained in the technical specifications. NRC officials point out that the amount of resource time devoted to meeting their own publication raquirements in the Federal Register would seem to f ail a cost / benefit analysis.
The administrative log jan which has been established has had the overall effect that processing of the most routine uncomplicated licensing amendments now nomally a requires a minimum of 90 to 120 days to accomodate the reviews and pre-noticing required by the new regulations. More complicated amendments may require six months or longer for review and approval. Additional discussion on this topic is found in Appendix B.
The changes currently suggested by the AIF Subcomittee make no direct effort to alter "the Sholly Process". It is the Subcomittee's conclusion that suggested cnanges will significantly reduce the volume of the technical specifications.
Futhemore, many of these items moved from technical specifications are precisely those which have generated trivial license amendments (for example: minor changes to lists or fuel-cycle dependent variables).
D. Increasing plant performance / availability The performance statistics of comercial U.S. reactors indicate that improvements should be made. A number of f actors have been reviewed to determine their impact on plant availability. Technical specifications have been shown to have considerable impact on plant operations. An NRC contractor review (SAIC evaluation for NRC-TSIP, " Impact of Technical Specifications on Operating Reactors In 1984")
reported that in 1984 alone, a total of 77 shutdowns were required by technical specifications. The subject of shutdowns contributed over 11/2% unavailability to the total industry perfomance. Industry estimates are higher than this value since the evaluation did not include start up delays, trips attributed to less flexible operating or test configurations nor normal maintenance personnal work diversions due to testing assignments. .The recort did. however, note that technicaljpecifications do also contribute 17%_to the forced outage _ hours. From theTtility perspectTve ...foVced outage hours... may nave greater impact on plant
~
operation since they are unplanned or unavoidable unavailabilities. Excessive testing at power required by current technical specifications contribute to higher scram frequencies. Owners group work (WCAo-10271 and NEDE-30851P) have addressed this and have quantified this-impact for the reactor protection systems. Many shutdowns were f or conditions of negligible risk. Some shutdowns may have placed a greater challenge on plant safety systems than if the plant had continued to operate while repairs were made. In sumary, technical specifications do contribute to unnecessary shutdowns.
IV. A New System of Technical Specifications In Section I, and more fully in Appendix A, the regulatory basis of technical specifications is reviewed. The current codification of the Atomic Energy Act requirement to include technical specifications in facility licenses,10 CFR 50.36, is deficient in at least one principal characteristic. While it provides a categorization of technical specifications' content, it provides no meaningful criteria regarding appropriate scope. This point was noted by the ASLAB in the Trojan proceeding and was the purpose of the 1980 Advanced Notice and 1982 Proposed R ul e. The 1982 Proposed rule remained deficient in this respect since it simply
- acknowledged that the technical specifications contained items of varying importance but provided no objective tests by which potential subjects could be judged. Subsection C, below, provides the missing criteria.
An additional unfortunate circtnstance has involved the concept widely held that ttchnical specifications are the only set of enforceable technical requirements that can be relied upon. This concept is embodied in NRR office letter No. 34 as well as indirectly in many other arenas. This should not be the case. Both the Enforcement Policy and history draw attention to many other enforceable
~
requirements. To be sure, the technical specifications do provide a set of ,
important requirements. However, many items of minor significance are included and others are appropriately not included. This proposal, when implemented, will allow the relocation of many items currently found in technical specifications. Thh should not be viewed as erradication_,. It is, rather, an appropriate restrucluring of a whole body of reguiltTory requirements.Section V will address the generic aspects of this restructuring and acknowledge that appropriate controls will be maintained.
This restructuring is diffe ent than that envisioned in the 1932 rule, in that no new major programs (Supplemental Specifications) are reconenended. This is premised upon the existence of other sufficiently controlled programs.Section V provides a detailed development of the general concepts of the proposed restructured system.
A. The General Philosophy of Technical Specifications The intent of technical specifications is to specify the limiting conditions for cptration to maintain the plant in a configuration during normal operation such that, if an accident or malfunction occurs, there is a high degree of assurance that the plant can be successfully brought to a safe. shutdown state.
B. Selection Criteria for the Present Technical Specifications PrIsent technical specifications ' consist of six sections. These are: Definitions, Safety Limits and Limiting Safety Systems Settings, Limiting Conditions for Operations (LCO), surveillance requirements, design features, adninistrative controls. The selection criteria in Section IV.C. has been developed principally for use with those items listed in the LCO section of present technical sptcifications. The selection criteria is not intended for use with the other sections of present technical specifications. These other sections have varying degrees of level of importance and in many cases repeat infonnation that is presently contained in the Final Safety Analysis Report or 10 CFR. The following provides the reconnendations for handling these other sections of present technical specifications.
i 12 Definitions - Retain in new technical specifications only those which provide clarification or have a special meaning within the document and which are not in -
general use.
Safety Limits, limiting Safety System settings - Retain in new technical specifications.
Surveillance Reouirements - Relocate from technical specifications. Surveillance requirements for items listed in the new technical specifications should be located in documents not controlled by the license amendment process. Surveillance, per
- se, is not of the same level of importance as the equipment involved. It is not a question of whether or not surveillance will be performed; but the details associated with surveillance, frequency and methodology, may be more effectively controlled by a program with an appropriate administrative control process.
Design Features - In many cases, infonnation contained in this section is redundant to that whicn is contained in the FSAR. Any additional details should be relocated into the FSAR. Items described herein that possess a high level of importance will have LCO's established for them. The design details provided by this section are not considered to be as high a level of importance. They can be effectively ,
controlled, and in many cases already are, by documents which are administratively controlled pursuant to 10 CFR.
Administrative Controls - In many cases, information contained in this section is redundant to that which is contained in the FSAR and 10 CFR. The administrative details contained therein are not considered to be as high a level of importance, except for those for which regulations exist (e.g., shif t staffing, non-routine event reporting). Any additional details can be relocated to the FSAR. These details can be effectively controlled in docunents which are administratively controlled pursuant to 10 CFR.
C. Selection Criteria For the Present Limiting Conditions of Operations (LCOs)
As a means of selecting those items covered by the present LCO's which are of greatest importance to plant safety and which should be included in the technical specifications, criteria have been established. A brief background as to how these criteria are selected is presented as a prelude to the criteria. A complete discussion of the development of the criteria is provided in Appendix D.
It is important that the plant be operated under conditions (pressure, power, water level, temperature, etc.) which are consistent with the safety analyses that have been performed. Each selected event has been evaluated to determine systems and limits which are essential to avoiding unacceptable results. The design basis analyses ' consider the potential initiating causes of threats to fuel and the nuclear system process barriers. Design basis accidents are those hypothetical events that potentially affect one or more of the radioactive material barriers and that are not expected during plant operations. The effects of these hypothetical cvents are analyzed giving consideration to plant conditions (loss of off-site power, single active f ailures, etc.) to examine events that result in the potential release of radioactive material. These are the established " bounds" of normal plant operation within which the conclusions of the safety analysis report are expected to remain valid. A fundamental purpose of technical specifications is to define and preserve the validity of the results and conclusions of the design basis ;
accident (DBA) analyses in the modes for which they are analyzec. '
The NRC, on a selective basis, has reviewed and accepted changes in the technical specifications using probabilistic assessment methodologies without changing the DBA analyses. These criteria are meant to be flexible enough so as not to preclude the future use of PRA methodology in establishing technical specifications.
The three criteria listed below ensure that the most important items' are maintained in the technical specifications. If an item meets any one of the three c'riteria, it should be retained in the technical specifications.
- Criterion 1: 'An installed system that is used to detect, by monitors in the control room, a .significant abnormal degradation of the reactor coolant pressure boundary, or; DISCUSSION: A basic concept in the protection of the public health and safety is i
the prevention of accidents. Systems are installed to detect significant abnormal degradation of the reactor coolant pressure boundary so as to' allow operator actions to either correct the condition or to shutdown the plant safely, thus reducing the likelihood of a loss of coolant accident.
This criterion is intended to ensure that technical specifications control those
- systems that detect excessive reactor coolant system leakage. Two specific examples of systems which are selected using Criterion 1 are:
Secondary system radiation monitors Reactor Building sump level instrumentation Criterion 2 A process variable that is an initial condition of the Design Basis Accident Analysis, or; DISCUSSION: Another basic concept in the protection of the public health and safety is that the plant shall be operated within the bounds of the initial conditions assumed in the existing Design Basis Accident (DBA) analysis. These analyses consist of postulated events, analyzed in the Final Safety Analysis Report (FSAR), for which a structure, system, or component must meet specified functional goals. These analyses are contained in Chapters 6 and 15 of the FSAR (or cquivalent chapters) and are identified as Condition II, III, or IV events (ANSI N 18.2) (or equivalent) that either assme the f ailure of or present a challenge to the integrity of a fission product barrier.
Process variables are parameters for which specific values or ranges of values have been chosen as reference bounds in DBA analyses and which are monitored and controlled in actual operation such that process values remain within the analysis bounds.
The purpose of this criterion is to capture those process variables that have initial values assmed in the DBA analyses, which are monitored and controlled. So long as these variables are maintained within the established values, risk to the public safety is presumed to be acceptably low.
Implicit in this criterion is the associated installed control room instrumentation that monitors and/or controls the selected process variable. Two specific examples of process variables selected using Criterion 2 are:
Movable Group Assembly Rod Insertion Limits Reactor Coolant System Pressure Limits
._s- , - . _ _ - ,.
14 1
Critecton 3: A structure, system, or component that is part of the primary success path of a safety sequence analysis and functions or actuates to mitigate a Design -
Basis Accident.
DISCUSSION: A third concept in the protection of the public health and safety is that in the event that a postulated DBA should occur, structures, systems, and components are available to function or to actuate in order to mitigate the consequence of the DBA. Safety sequence analyses or equivalent have been performed in recent years and provide a method of presenting the plant response to an acci dent.
A safety sequence analysis is a systematic examination of the actions required to mitigate the consequences of events considered in a plant's DBA analysis, as presented in Chapters 6 and 15 of the plant's Final Safety Analysis Report. Such a safety sequence analysis considers all applicable events, whether explicitly or implicitly presented. The primary success path of- a safety sequence analysis consists of those actions assurned in the design basis accider.t analysis which limit the consequences of the events to within the appropriate acceptance criteria.
It is the intent of this criterion to capture into technical specifications only ,
. those structures, systems, components that are part of the primary success path of l a safety sequence analysis. Implicit in this criterion are those support systems that are necesssary for items in the primary success path to successfully f uncti on. The primary success path is equivalent for each DBA to the combinations and sequences of Equipment assumed to operate when responding to the event which results in acceptable plant acc.ident response (including consideration of the single f ailure criterion).
Two specific examples of struct~ures, systems, and components which are selected using Criterion 3 are; Reactor trip system instrumentation Primary system safety valves D. Disposition of Requirements not Appropriately Contained in Technical Specifications Those structures, systems, components, process variables and administrative sections that would no longer be in the technical specifications would be placed in other documents (FSAR, QA Plan, Fire Protection Plan, procedures etc.). It must b2 recognized that there are other means of validating the overall safety of the plant such as the design review (FSAR, SERs etc), the QA plan, the operating procedures, etc. These other documents have regulatory controls placed on them and are both inspectable and enforceable. For example, action statements are presently found in operating and emergency procedures. There is information presently found in other documents, for example the emergency plan, that have equivalent importance in protecting the health and safety of the public but are not located in the-technical specifications. A detailed discussion of other documents available, their present control and a suggested process for controlling those items no longer in technical specifications is provided in Appendix F.
l 1
I l
j
'~
l .
V. IMPLEMENTING THE NEW SYSTEM The infomation. contained in this section describes the process for implementing the improved technical specifications and companion docments. It-discusses the changes to existing regulations, and changes to regulatory guidance documents. It also describes the conversion of both Standard Technical Specifications and custom technical specifications by application of the new concept. -
A. Applicability of the New System
- One of the key industry premises under which the improvments in technical specifications was initiated was that conversion to the new concept would be v31untary for plants having either custom technical specifications or Standard Technical Specifications. Additionally, conversion to Standard Technical Specifications should not be a prerequisite to obtaining the benefits of this approach. With this understanding, the suggested criteria, suggested changes to rcgulations and control process are applicable to custom as well as Standard Tcchnical Specifications.
B. Changes to the Regulations .
The selection criteria for detemining what should be in technical specification should be codified in the regulations. As discussed in Section IV.C. of this -
document, those LCOs to be located in technical specifications define and preserve the validity of the results and conclusions of the design basis accident analyses.
They assure that the f acility is maintained in such a configuration as to maximize the probability of a successful and safe recovery from the accident.
4 Therefore, recognizing the level of importance to safety of these limitations, the criteria used to detemine these limits should be made part of the regulations.
The suggested changes to 10 CFR 50.36 are discussed in Appendix 11.
C. Short Term Improvements There are improvements in the present technical specifications that can be made now without awaiting the results of rulemaking. These include, but are not limited to (1) removal of lists such as containment penetration lists, fuse lists, etc. (2) the removal of requirements that are already in the regulations and which must be cet.by the licensee, and (3) attaining a clearer understanding of the interpretation of the term, " operable". These are discussed individually in
- Appendix G. The industry encourages the NRC to address these suggested
- improvements now and obtain the benefits as soon as possible. The AIF Subcomittee will continue to interact with the NRC on these and additional generic improvements.
D. Conversion of Technical Specifications The conversion process for licensees, whether their current technical i
specifications are custom or standard, has two primary characteristics: (a) itis voluntary; and (b) it does not seek to bypass appropriate backfit controls, Regarding the potential for backfits, it is expected that the. general approach will be for each Owners Group to collectively review the applicable technical specifications. Each licensee will have the benefit of such partitioning efforts as they develop their individual optimization proposals. Sinc? the plant is '
presently operating with an acceptable set of technical specifications which are part of the license, certification of the revised technical specifications should '
not be required. l i
l l
Any potential future additions to technical specifications, whether identified in cptimization effort (s), identified by the NRC staff in their review of generic or -
plant-specific issues, or otherwise, will need to be judged against the criteria presented in Section IV C above and against appropriate cost-benefit reviews expressed in 10 CFR 50.109.
E. Format and Style of New Technical Specifications .
The contents of new technical specifications are established by the selection criteria previously described. In order to make the new technical specifications a
- core user friendly document, certain enhancements to the style and format should be i ncorporated. Some areas of present technical specifications that tend to make the document difficult to use include extensive equipment listings, cycle dependent
~
variable values, and single systems referred to in several specifications with overlapping or contradictory requirements.
~
In order to correct these and other style and ferrat I"oblems, it is proposed that an industry standard writers guide for technical spc;ifications be developed.
There presently exists an industry star dard (ANSI /ANS - 58.4 -1979, Criteria for Technical Specifications for Nuclear Poser Stations) that could be revised to ,
include not only the selection criteria previously provided but also guidance in the preparation of technical specifications. This guidance would be in the form of a witers guide to ensure that writers produce technical specifications that are readable, complete, convenient, accurate and useable to control room personnel.
Such a guide would include suggested format principles as examples for use by citers in preparing the document. It would also include a discussion of style and sentence structure that would empnasize the use of concise directions. Further, it would provide guidance to allow information to be presented in a simple, f amiliar, and unambiguous manner. Consistency in styie and sentence structure of technical specifications will improve the ability to understand and use them just as it has been found to improve the effectiveness of revised emergency operating procedures.
This effort will be initiated shortly and will build upon the writing criteria presently detailed in the ANS standard. The resultant document will be a revised ANSI /ANS 53.4 standard which can be used by the nuclear industry as guidance in the preparation of new technical specifications. Further, NRC could adopt this standard by issuing a generic letter or Regulatory Guide to this effect.
A-1 l Appendix A: Regulatory History and Current Regulatory Requirements and Guidance As used in the nuclear industry, the tem " technical specifications" originates in Szction 182a of the Atomic Energy Act of 1954, 42 USC 2232(a). That Section 1822 4
provides in pertinent part that:
"In connection with applications for licenses to operate production or utilization f acilities, the applicant shall state such technical specifications, including information of the amount, kind, and source of j special nuclear material required, the place of the use, the specific characteristics of the f acility and such other information as the Comission may, by rule or regulation, deem necessary in order to enable it to find that the utilization or production of special nuclear material will be in accord with the comon defense and security and will provide adequate protection to the health and safety of the public. Such technical specifications shall be a part of any license issued."
This statutory directive has been implemented in Section 50.36 of the Comission's regulations in 10 CFR Part 50, entitled " Technical Specifications." Before 1968, ,
1 Section 50.36 required technical specifications to include "those signi. icant design features, operating procedures, and operating limitations which were considered important in providing reasonable assurance that the f acility (would) be l
constructed and operated without undue hazard to public health and safety."
Technical specifications that were formulated in accordance with this regulation, as it was then written, generally contained more detailed design information than was considered to be necessary to assure safe reactor operation. These technical specifications proved to be difficult to organize, unduly restricted flexibility of reactor operation, and necessitated the processing of many changes that were not
, significantly related to safety (47 FR 13370).
In December,1968, the Atomic Energy Cortnission (AEC?, predecessor of the NRC, amended its regulations in Sections 50.36 and 50.59 (33 FR 18612). Section 50.36 c:as amended to include a more precise definition.of those categories of technical specifications that must be included in an application for an operating license.
The amended regulation narrowed the scope of the material contained in technical
! specifications by defining five specific categories of technical specifications.
! The five categories defined for nuclear reactors are: (1) safety limits and i
limiting safety system settings, (2) limiting conditions for operation, (3) surveillance requirements, (4) design features, and (5) administrative controls i
(47 FR'13370).
The Statement of Consideration which accompanied promulgation of 10CFR50.36 in its present forn made specific reference to the then recently issued " Guide to Content i' of Technical Specifications for Nuclear Reactors" (November 1968, 33 FR 18610).
That guide spoke of technical specifications in terms of " conditions governing c j operation of a f acility that cannot be changed without prior Comission approval" and that represent " legal bounds within which the licensee is required to operate -
the f acility." It went on to state the technical specifications "related to j technical matters should consist of those features * *
- of the f acility that are i
of controlling importance to safety; the identification of such features was to be accomplished by thorough safety analysis of the f acility, the analysis being based on current knowledge and understanding of safety needs and techniques."
l A-2 In 1979, the Atomic Safety and Licensing Appeals Board clarified both the statutory and regulatory underpinnings for techqical spect'ications when it upheld a prior licensing bo.srd decision concerning an application to expand the spent fuel pool at the Trojan Nuclear Plant (Portland General Electric2 Co., et al. (Trojan Nuclear Plant), ALM-531, 9 NRC 263 (1979)). At issue w H a contention by'the fitate of Oregon that certain items from the design report on the spent fuel pool's expansion should be given the status of technical specifications to the operating license.
This contention was denied and the Appeals Board held that water chemistry limits for minimizing corrosion to fuel elements and spent fuel racks "need not be carried over into a technical specification to insure a margin to safety." In its ruling.
the Appeal Board wrote:
"From the foregoing it seems quite apparent that there is neither a statutory nor a regulatory requirement that every operational detail set forth in an applicant's safety analysis report (or equivalent) be subject to a technical specification, to be included in the license as an absolute condition of operation which is legally binding upon the licensee unless and until cha%ed with specific Co rtission approval. Rather, as best we can discern it, the contemplation of both the Act and the regulations is that technical specifications are to be reserved for those matters as to which ,
the imposition of rigid conditions or limitations upon reactor operation is deened necessary to obviate the possibility of an abnormal situation or event giving rise to an immediate threat to the public health and safety".
On July 8,1980, even before the Appeal Board's ruling in the Trojan case, the Ccmission published an Advance 'iotice of Proposed Rulemaking (APR) (45 FR 45915),
requesting coments on the desirability of changing its regJlations on technical specifications to: (1) establish a standard for deciding which iteas derived ' rom the safety analysis' report must be incorporated into the technical specifications l for a facility; (2) modify the definitions of categories of technical spccifications to focus more directly on the aspects of reactor operation that are
, important to the protection of the health and safety of the public; (3) define a new category of requirements that would be of lesser imediate importance to safety than technical specifications, thereby providing greater flexibility to both the NRC and licensees in processing proposed changes; and (4) establish appropriate conditions that must be met by liceqsees to make changes to the requirements in the new category without prior approval (47 FR 13370-13371). (Ccments received in response to the AP R were strongly in favor of a rule change to incorporate these concepts, see 47 FR 13371 and SECY S1-672.)
On March 30, 1932, af ter considering the public coments made in response to its previous APR, the NRO published a proposed rule for ccment (47 FR 13369). This is the so-called " George" rule to split the technical specifications, relocating some material to a document called Supplementary Specifications. In the supplementary information portion of its proposed rule, the Ccmission explained (Id. at 13370) that the recent Trojan case "... highlighted the need to establish sp;cific criteria in the regulations for deciding which items derived from the safety analysis report must be included in the technical specifications incorporated in the license for a f aellity." The Comission went on to explain the then current problems with technical specifications:
"In addition, the substantial growth in both the number of items and in the detail of the requirements contained in technical specifications that has taken place since the STS were instituted indicated that more precise
A-3 definitions of the existing categories of technical specifications contained in Section 50.36 are needed. The Commission is concerned that the increased volurne of technical specifications lessens the likelihood that licensees will focus attention on matters of more immediate importance to safe operation of the f acility.
While each of the requirements in today's technical specifications plays .a role in protecting public health and safety, some requirements have greater imediate importance than others in that they relate more directly to f acility operation.
These are the requirements that pertain to items which the f acility operator must b2 aware of and must control to operate the facility in a safe manner. To a large cxtent, the relative importance of these requirements, as distinguished from those related to long-term effects of concerns, may have been diminished by the increase .
in the total volume of technical specification requirements.
~
Moreover, increased volume and detail of technical specifications and the resultant increase in the number of proposed change requests that must be processed have increased the paperwork burden for both licensees and the NRC staff. This is because Section 50.36 requires that technical specifications be included in each Cperating license; thus, any proposed change, regardless of its importance to ,
safety, must be processed as a license amendment. For changes involving matters of lesser importance to safety, the processing of a license amenhent with the associated increased paperwork has had no significant benefit with regard to protecting the public health and safety".
A draft Comission paper containing a proposed final rule was issued in a March 21, 1983 NRC memo from Darrel Eisenhut (therr Director, D0L) to other division directors t:ithin NRR. This draf t Comission paper explained that 27 out of 29 comentors had cxpressed support for the general concepts of the 1982 proposed rule.
This rulemaking for technical specification reforms was never completed. 'The Commission suspended efforts on the rule in sometime around mid-1983 in order to concentrate staff resources on the issuance of near term operating licenses.
The concept for technical specification refonn outlined in the 1982 proposed rule and also in the 1983 draf t Comission paper consisted of dividing the existing tcchnical specifications inte two general categories, each with corresponding provisions regarding the NRC's treatment of licensee-initiated changes. Basically, the proposed system would have required prior NRC approval of a license amentent for changes to certain items still called technical specifications in the 1982 proposed rule. These items would include such things as process-variable safety lir.its, limiting safety system settings, operational limits and conditions, certain check and test requirements, and operating shif t crew and staff composition requirements. Items in the other category, called supplemental specifications, c:uld be changed without prior NRC approval or license amen &ent since they would n:t be considered directly a part of the operating license. For such changes, however, NRC would have to require licensees to prepare and report safety evaluations on an at-least-annual basis. Also, the NRC Regional A kinistrator could quickly revoke such changes in writing, if he so desired. The items in this second category would include such things as operating state and status of systems; provisions for monitoring, inspecting, testing, and calibration; and akinistrative provisions related to licensee management practices at the facility. Although the supplemental specifications category woJ1d not have been directly a part of the license, it would nevertheless have been " linked" to the license via a license condition in 10CFR50.54.
_ _ _ _ . a_ - a .- _n_w_.;L ,J we.
A-4 Finally, the 1982 proposed rule would not have required operating plants to implement the proposed two-category system. Instead, for these plants, implementation would have been voluntary. Indeed, the Comission explained in its notice that " Technical Specifications issued before that date 180 days before the final rule's effective date would not be required to be changed; however, upon request by a_ licensee to convert the existing technical specifications to' the new scope, content (emphasis and 47 added). fomat}3374.the FR NRC would take action to grant the request" Also, the proposed rule did not explicitly
. condition these plants' use of the two-category system on their adaption of Standard Technical Specifications as a prerequisite. The Regulatory Impact Analysis accompanying the 1932 proposed rule, however, suggested the opposite.
The regulatory basis for technical specifications is embodied in 10 CFR 50.34(b)(6)(vi) that requires, as part of the Final Safety Analysis Report (FSAR),
each application for a license to operate a facility, provide proposed technical specifications and bases or reasons for such technical specifications in accordance uith the requirements of 10 CFR 50.36. Under 10 CFR 50.36, technical specifications are incorporated into licenses authorizing operation of a production or utilization f acility. According to 10 CFR 50.59, after issuance of an operatin
- license, the Technical Specifications can only be changed with prior NRC approval,g which is granted by issuance of an amendnent to the license. Additional regulations related to technical specifications are listed in Table A-1.
In addition to regulations, supplemental guidance has been issued over the years t for the preparation and content of technical specifications. Chapter 16 of the endorses the use of Standard Technical Standard Specifications Review for plants Plan (NUREG-0800)ly inEach current licensing.
NSSS vendor has a set of such Standard Technical Specifications for its system. Regulatory Guide 1.16,
" Reporting of Operating Infomation--Appendix A, Technical Specifications,"
presents an acceptable basis for reporting operating infomation as required in the 4
technical specifications. These include startup reports, annual operating reports, and monthly operating reports, all of which are considered routine reports. In addition, NUREG 1022 provides infomation on reportable events.
ANSI /ANS-58.4, " Criteria for Technical Specifications for Nuclear Power Stations,"
provides a detailed conceptual framework which can be applied to the preparation of technical specifications, criteria for selecting subjects and values to be included in the technical specifications, and criteria for developing technical specification bases. The attached Table A-2 lists additional material which provides guidance for or insight into techncal specifications.
i
l A-5
_ Table A-1 1 Primary Regulations for Technical Specifications 10 CFR 50.36, " Technical Specifications."
Describes information to be included in technical specifications, including items in the following categories:
- 1. Safety limits, limiting safety system settings, and limiting control settings
- 2. Limiting conditions for operation 3.~ Surveillance requirements
- 4. Design features
- 5. A tinistrative controls 10 CFR 50.36a, " Technical specifications on effluents from nuclear power reactors."
Requires technical specifications that will provide for keeping releases of radioa:tive materials to as low as is reasonably achievable.
10 CFR 50.59, " Changes, tests and experiments."
Indicates that any changes in technical specifications must be by license menhent.
Secondary Regulations Governing Aspects of Technical Specifications 10 CFR Part 2. Appendix C, " General Policy and Procedure for NRC Enforcement Actions."
Although this is a general policy statement, rather than a specific regulation, it is regarded as applicable. In its discussion of the varying degrees of safety, safeguards or environmental significance that regulatory requirements have, a footnote indicates that the term " requirement" includes, among other items, technical specifications. In determining severity categories, technical specifications are specifically mentioned.
10 CFR 50.46, " Acceptance criteria for emergency core cooling systems for light water nuclear power reactors."
Technical specifications are required to conform to the requirements of this s ecti on.
10 CFR 50.48, " Fire protection."
Provides for fire protection features to be completed in accordance with the technical specifications.
A-6 10 CFR 50.54, " Conditions of licenses."
Ties minimum shif t requirements to definition of operational modes as given in the technical specifications; indicates no change can be made frcrn techncal specifications as incorporated into a license; provides for ceparture from technical specifications in an emergency.
10 CFR 50.55a, " Codes and standards."
If an inservice inspection program revision conflicts with the technical specifications, the technical specifications are to be revised to conform to the ISIS.
10 CFR 50.71, " Maintenance of records, making of report."
Maintenance of records required by technical specifications.
10 CFR 50.72, "Imediate notification requirements for operating nuclear power reactors."
10 CFR 50.73, " Licensee event report system."
10 CFR Part 50, Appendix I, "Neerical guides for design objectives and limiting conditions for operation to meet the criterion "as low as is reasonably achievable" for radioactive material in light-water-cooled nuclear power reactor effluents."
10 CFR Part 50, Appendix J, " Primary reactor contairrnent leakage testing for water-cooled power reactors."
Provides test requirements to assure that allowable leakage rates in technical specifications are not exceeded; provides guidance for establishing appropriate contairrnent leakage test requirements in technical specifications.
10 CFR Part 50, Appendix K "ECCS evaluation models."
Provides guidance on acceptable evaluation models using the maximtsn peaking f actor allowed by the technical specification.
10 CFR Part 50. Appendix H, " Reactor vessel material surveillance program requi rements."
Indicates a change in the technical specifications must be made if mandated by.
the pressure-temperature limits.
10 CFR Part 50, Appendix R, " Fire protection program for nuclear power f acilities operating prior to January 1,1979."
Refers to Standard Technical Specifications for definition of hot standby and hot shutdown.
A-7
- 10 CFR Part 55.22 " Content of senior coerator written examination."
Need for operators to know design and operating limitation in the technical
! specifications.
10 CFR Part 55, Appendix A. "Requalification programs for licensed operators of production and utilization f acilities." , ,
Technical specifications are to be included in any requalification program.
10 CFR 70.32, " Conditions of licenses."
Any change which would decrease the effectiveness of the plan for physical i protection of special nuclear material, as incorporated in technical specifications must be submitted for a license anendment.
I 10 CFR 72.16. " Contents of Application: Technical Specifications" l
- Sets forth technical specification requirements for independent spent fuel i storage installation.
I t
i l
i 1 e
l i
j I
i l
j A-8 Table A-2 Additional Guidance for Technical Specifications ANSI /ANS-58.4 Criteria for Tech Specs for Nuclear Power Stations
{ GENLTR-82-16 NUREG-0737 Tech Specs - PWR's i
i 3
. GENLTR-83-02 NUREG-0737 Tech Specs - BWR's !
GENLTR-83-13 HEPA Filters-Standard Tech Specs GENLTR-83-26 D/G Fuel Oil Requirements GENLTR-83-27 Surveillance Intervals in Standard Tech Specs l
- GENLTR-83-30 D/G Testing GENLTR-83-35 NUREG-0737 Tech Specs
- l i
GENLTR-83-37 NUREG-0737 Tech Specs GENLTR-83-43 Reporting Requirements of 10 CFR 50.72 and 50.73 and Standard Specifications
] GENLTR-84-13 Tech Specs for Snubbers i GENLTR-84-15 Proposed Staff Actions to Improve and Maintain biesel Generator
] Reliability 1 NUREG-0103 Standard Tech Specs - Babcock & Wilcox i
NUREG-0123 Standard Tech Specs for General Electric Boiling Water Reactors l NUREG-0133 Preparation of Radiological Effluent Tech Specs for Nuclear Power Plants
.NUREG-0212 Standard Tech Specs for Combustion Engineering Pressurized Water Reactors
- NUREG-0452 Standard Westinghouse Tech Specs l NUREG-0472 Draf t Radiological Effluent Tech Specs for PWRs 1
} NUREG-0473 Radiological Effluent Tech Specs for BWRs j NUREG-0839 A Survey by Senior NRC Management ;
j NUREG-1024 Tech Specs-Enhancing the Safety Impact NUREG/CR-3082 Probabilistic Approaches to LCOs and Surveillance Requirements for Standby Safety Systems RG-1.16 Reporting of Operating Infomation-Appendix A Tech Specs j C-4.8 Enviromental Tech Specs for Nuclear Power Plants i ,
I
-- - - . -.._ _ . - ._ - . - - _ --_ .-_ _. . .- =.-
A-9 '
l j
i RG-10.1 Compilation of Reporting Requirements for Persons Subject to NRC i Regulations l SECY-82-003 10 CFR 50.73 Establishing the Licensee Event Report (LER) System t
SECY-83-487 Revised General Statement of Policy and Procedure for Enforcement
- Actions
{ . SRP-2.4.14 Tech Specs and Emergency Operation Require.nents i
SRP-16.0 Tech Specs I
I 1
.i ,
i i
i i
l t
I i
f I
I i !
l t
4 f
1 t
. I I l i
i j '
. B-1 1
Appendix B: The Paperwork Burden of the Sho11gequi,ations, .
l B ackgroun_d 4
At one point during the TMI-2 accident, it became necessary to vent the containnent j
atmosphere to the local surroundings. At the request of the licensee, the NE approved a license amentent that allowed the venting which would have normally
- b;en outside existing operating requirements. Steven Sholly, a participant in the TMI-1 restart hearings, brought suit against the NR
- in 1980 because the licensee
- of TMI-2 was allowed to release some gaseous radioactivity into the atmosphere i t:ithout local residents being perinitted meaningful participation through the hearing process prior to the release. In Sholly vs. NRC a 0.C. Circuit Coart ruled
~
. that a prior hearing is required if reqacsted even iTc'a'ses where the NRC has detennined that the subject of the proposed amenhent to the license involves "no
. significant hazards" consideration.
I i
The specific regalations resulting from the Sho11y decision are contained in 10 CFR 50.91 and 10 CFR 50.92. They require that, for any license anenchent, the licensee /NRC are obliged to take three actions: 1) make notice in the Federal ,
Register for public ccmnent 2) initiate consultations with the applicable state's
, designated official and 3) make a finding of "no significant hazard" l consideration. The naclear indJstry is concerned that these obligations: 1)cause
) an unnecessa y psperock bJeden for all parties due to Federal Register noticing, i form memos, evaluations, etc., 2) that the opportunity for hearing will be abused i
by opponents o' nJclear p m r in trivial license anendnent esses, 1) that res:htng j a finding of "no significant hazard" consideration is subject to ma9y levels o'
. scrutiny.
Discussion i
Although the Sholly regJlations are applicable to all license amenheits, the i
purpose of this discussion is to address the impacts of the Sholly regulations on catters related to technical specifications initiated license amenhents. The
, primary connection is by way of changes, tests and experiments as described in 10 CFR 50.59. This regulation requires that any proposed change, test or experiment t be evaluated to determine whether or not it involves an unreviewed safety question (USQ) or a c5ange to the techq1 cal specifications. If the change, test or ;
! experiment does involve a USQ or a technical specification change, prior Nt approval and a license amenheit is required. Although 10 CFR 50.59 provides
! criteria that identifies USQs according to degree of impact on FSAR and other licensing bases, there is no such criteria provided to evaluate a technical j specification change. In other words, any change to the technical specifications, 2
however slight, assitnes the $1ne degree or severity in regard to the projected j adverse impact on health and safety of the public as a USQ. In regard to Itcense
- am
- ndment aspects,10 CFR 53.91 and 10 CFR 53.92 (i.e., Sholly regalations) rea; ire that notice be published in the Federal Register, the State be consulted and that
no significant hazard" consideration be deterntned for all license snendnents.
l .
( The process of amending many items contained in current technical specificatio1s now imposes an unnecessary, costly, and useless burden on the N C and Itcensees to
, ceet administrative requirements which seem not to serve the needs of the public or
- th? nuclear indJstry.
2 4
l
B-2 Under the current amen &ent system implemented in 1983, the Commission requires an applicant requesting an amen h ent to its operating license or revision to technical specifications to provide its appraisal on the issue of whether or not the amen hent involves a significant hazards consideration using the standards in
, 10CFR50.92; and if the amen &ent involves the emergency or exigency provisions, to address the features upon which the Comission must make its findings. The licensee also is required to notify the State authorities of the.anentent and the results of its significant hazards consideration review.
- When the Comission receives the amen &ent request, it first decides whether or not
. there is an emergency or an exigency. If there is no emergency, it then makes a preliminary decision, called a " proposed detemination" about whether the amentent involves a "no significant hazards" consideration. At this stage, if the Commission decides that no significant hazard consideration is involved, it can i issue an individual Federal Register notice or list the amen &ent in its next bi-conthly publication in the Federal Register. This bi-monthly publication lists not only the amentent requests received for which the Comission is publishing notice under 10 CFR 2.105 to provide opportunity to request a hearing, but also i provides a reasonable opportunity for public coment by listing this and all t
amentent requests received since the last such bi-monthly notice, providing a brief description of the amen &ents and the f acilities involved, noting the
- proposed "no significant hazards" consideration deteminations, soliciting public
, coment on the deter ninations, and providing for a 30-day coment period. The notices are lengthy and sometimes difficult to prepare because of their judgmental character. While awaiting public coment, the Comission proceeds with the safety analysis, and the State in which the licensee is located is consulted on the
- Comission's proposed determination of "no significant hazards".
Af ter the public coment period, the Comission reviews the coments, considers the safety analysis, and reaches its final decision on the amenhent request. If it
. decides that "no significant hazard" consideration is involved, it can publish an individual " notice of issuance" or publish the notice of issuance in its system of bi-conthly Federal Register notices and thus close the public record. Note that the Ccrrnission would not have to make and publish a final determination on its "no
- significant hazards" consideration for most cases because such a detemination is l
needed only if a hearing request is received and the Comission decides to make the am:nhent imediately effective and to provide a hearing af ter issuance rather than before.
The 1993 Sholly legislation has greatly increased the amount of paperwork involved for both the utility and the NRC in issuing license snenhents.
These complications include such things as: internal controversy between technical staff and legal staff which could necessitate extensive rewrites and meetings,
- disagreenents between NRC staff and licensees, extensive coments to the Federal Register notices, and opposition resulting in the need to hold a public hearing.
The public does not appear to benefit from these hearing opportunities for many items now included in the technical specifications. Many licensing amen &ents are sought to change corporate structure, or update equipment lists which are currently contained in the technical specifications. The atinistrative log jam which has been established has had the overall effect that processing of the most routine r uncomplicated Itcensing amen &ents now normally requires a minimum of 90 days to l
i
B-3 accomodate the reviews and pre-noticing required by the new regulations. Slightly i
tire complicated amendnents require six months or longer for review and approval.
The possibilities for abuse of the adninistrative process already has had an effect on utility initiatives to improve their nuclear stations since it reduces the utility's incentive to make such changes. A reduction in the scope of the technical specifications to those issues which are clearly important to the safe Cperation of the station reduces the risk the utility f aces in proposing tprovements, as well as minimizes unnecessary paperwork burdens for both licensees and NRC staff.
In the past, when the volume and level of detail routinely included in the technical specifications was much less than it is today, the impact of Sholly r;gulations would most likely have been much less. This is primarily due to the f act that as design evolution changed minor details in the plant, the technical specifications were written to such a general level of detail that no associated technical specifications changes would be required. However, since today's technical specifications contain information in a greater level of detail, minor plant changes may give rise to the need for a technical specification change; which as discussed above, potentially giving rise to the hearing process. .
Federal Register notices, state consultations and findings of "no significant hazard" consideration creates a wealth of paperwork. Table B-1 provides a list of new forms and reporting documentation that the Sholly regulations have generated.
Table B-1 is taken from NRC DLOP-228. It is not clear that this proliferation of paperwork could survive a cost / benefit analysis when reviewed from the aspect of positive impact on the health and safety of the public; especially in the case of minor changes to the technical specifications. That should be the bottan line .
criterion by which the amount of paperwork is judged.
In addition to the mendnent process itself, there is the paperwork burden associated with any hearing especially a protracted hearing. The paperwork
- consnitment to any hearing effort may be unestimatable.
In a letter from C.R. Anderson and John A. Van Wagenen of the Investigative Staff of the House Ccrrnittee on Appropriations to the Comission on March 15, 1985, it tas stated that, "The Sholly legislation has greatly increased the mount of paperwork involved in issuing license mendnents. The monthly Federal Register notice is typically 50 pages long. The notices are lengthy and sometimes difficult to prepare because of their judgmental character. Since the NRC Sholly rule went intosignificant no effect in May, 1983, hazard 1,625 notices have been issued for mendnents involving consideration. Only 14 public or state coments have been received on these detenninations, and only 15 hearing requests have been i received."
The changes currently suggested by the AIF Subconynittee make no direct effort to alter "the Shelly Process". It is the Subcomittee's conclusion that suggested changes will significantly reduce the volume of the technical specifications.
Futhennore, many of these items moved from technical specifications are precisely those which have generated trivial license amendments (for example: minor changes to lists or fuel-cycle dependent variables).
t
Table B-1 -
i r
- 1. Interim Final Rule - Notice and State Consultation
- 2. Interim Final Rule - Standards for Determining Whether License Amendments j Involve No Significant Hazards Considerations
- 3. Flow Diagram - State Consultation, Noticing and No Significant Hazards Consideration Procedures -
l 4 Initial No Significant Hazards Consideration Determination and Noticing Action (Fonn)
- 5. Sample Individual FR Notice of Opportunity for Prior Hearing' 4 6. Sample Individual Notice of Issuance of Amendment
- 7. Sample Monthly FR Notice
- 8. Sample Memorandum Requesting Input into Monthly FR Notice i 9a. Sample Individual FR Notice of Proposed No Significant Hazards Consideration .
- Determination and Opportunity for Hearing 9b. Sample Memorandam Requesting Input into Monthly FR Notice
- 10. Sample Memorandum to Regional Public Aff airs Officer and Press Release
- 11. Sample Memoranda: Requesting Input into Monthly FR Notice
- 12. Final NSHC Detennination (Form) i 13. Sample Individual Notice of Issuance of Amendment and Final Detennination of '
NSHC 14 Sample Memorandam Requesting Input into Monthly FR Notice f
- 15. Sample Memorandum Requesting Input into Monthly FR Notice
- 16. Sample Individual FR Notice of Consideration of Issuance and Proposed NSHC Detennination (Short Notice) 1 1
C-1 Apoendix C: Public Ccments Regarding the 1982 NRC Proposed Rulemaking on Technical Specifications The majority of comentors (27 of 29) supported the essence of the proposed rulemaking in that it would establish two separate categories which define those ,
specifications which are part of the operating license (OL) and of imedi, ate importance to safety and thus subject to direct NRC control and those specifications which are not part of the OL but include other items important to safety and under licensee control. However, there was disagreement among the
- comenters over suitable nomenclature for these categories. Some comentors i recomended that the new technical specifications be submitted as a draf t,18 months prior to the scheduled issuance of the OL and not be included in the final safety analysis report (FSAR) until issuance of the OL.
Several comenters expressed concern that implementation of the proposed rule on plants issued OL's relatively soon af ter the effective date of the final rule could adversely affect plants in the licensing process and therefore recomended that the effective date be extended from ISO days af ter publication to one year af ter publication. A few ccynmenters recomended that the proposed rule not become final
- and implemented until the existing Standard Technical Specification:; (STS) are revised to be consistent with the proposed rule and suggested that the proposed rule should not apply to the second unit of a two unit plant if the first unit is licensed with technical specifications developed before the effective date of the proposed rule.
I One comenter proposed that the rule allow licensees to request an amentient to the OL designating those sections of the OL that may be revised without prior NC approval provided the licensee complies with the provisions for changes to supplemental specifications. Several comenters suggested further that the Operational Specifications section of the technical specifications should contain only requirements that are of imediate importance to safety and under direct c:gnizance of the operator or that they be limited to the first three types of safety analysis assumptions:
- 1. Values of process variables that must be kept within certain bounds;
- 2. Operating state of equipment (e.g., valve position) that must be maintained;
- 3. Operating status (or operability) of equipment that must be maintained.
Comenters recomended revising the definition of " limiting safety system settings" to clarify that these are chosen so that the automatic protection actions they i initiate will prevent the violation of a safety limit during nomal operation and anticipated operational occurrences but not during accident conditions. One
, comenter stated that the proposed narrowing of the operational limits and 1
conditions category of the operational specifications to those items of an i imediate importance to safety should be rejected because in an unanticipated series of events, items not of imediate importance to safety could become of primary and imediate importance to safety while others felt that the definition of operational limits and conditions should more clearly define the manner in which an operational limit and condition is associated with the perfomance of the four
, safety functions:
i i
C-2 l
- a. controlling reactivity
- b. cooling the fuel
- c. protecting the integrity of fission product barriers ,
- d. limiting the release cf radioactive fission products following an accident
! Some comenters felt that the four safety functions were broad, vague and somewhat overlapping (e.g., (c) & (d)) and consequently one comenter suggested expanding them to ten critical safety functions:
- a. Reactivity Control
- b. Reactor Coolant System Inventory Control
- c. Reactor Coolant System Pressure Control
- d. Core Heat Removal
- e. Reactor Coolant System Heat Removal
- f. Contairrnent Isolation
- g. Containment Pressure and Temperature Control
- h. Combustible Gas Control
! 1 Indirect Radiological Release Control ,
- j. Maintenance of Vital Auxiliaries A few cocinenters stated that check and test requirements should be part of the
' supplemental specifications since all surveillance requirements can safely be part cf the supplemental soecifications provided corrective action requirements remain in operational specifications under the direct control of a licensed operator.
i i
Other comenters recomended that at least one check and test requirement correspond to each limiting safety systern setting. Some comenters suggested that the provisions for operational staffing and reporting requirements be either deleted entirely or included in the supplemental specifications while one comenter strongly recomended that the use and control of procedures (presently Section 6.8.1, 6.8.2, and 6.8.3 of the STS) be retained as operational specifications to assure uniform treatment of the procedural requirements.
\
I Several comenters recomended relocating the principal design feature specifications to supplemental specifications since they are not under direct i control of the operator while several others recomended deleting them entirely j since changes in design features will be tracked in FSAR updates and are subject to evaluation in accordance with 10 CFR 50.59 in addition to the reason stated above.
l Sane conenenters reconenended allowin licensees the convenience and flexibility of 1 maintaining supplemental specificat ons in a separate document from the FSAR and l that the method for changing these specifications be the same as for changing the l FSAR (10 CFR 50.59). A few comenters reeminended revisions to the text to make i.
C-3 absolutely clear that specifications should cover only primary safety concerns to l plant operations and to reduce or prevent expansion of the overall specifications.
One cuenter recomended eliminating the need for, or at least reducing the tineliness of Licensee Event Reports to events under the supplemental specifications and that the rule be written to reduce or eliminate enforcement actions under these specifications so that plant operators could focus their
- attention on matters important to safe plant operation.
! One comenter requested that the reporting requirements for f ailing to meet the l centrol provisions be carefully exmined to assure that they are appropriate for the circumstances and that the rule explicitly state that a change in plant i cperating mode is not required when a control provision is not met. Others recomended that requirements for control provisions be based on an evaluation '
thich attempts to optimize the relationship between safety needs, plant capabilities and power production and that time intervals associated with conitoring provisions be derived from a detailed study of system needs, safety needs, equipment considerations and the ability to perform the required surveillance. One comenter also recomended that the atinistrative provisions be deleted from the supplemental specifications and be allowed to exist in other forms and documents.
- One comenter stated that the provisions for changing suppicmental specifications involving a decrease in their effectiveness are excessively restrictive and that licensees should be given the opportunity to evaluate surveillance frequencies at their discretion and make changes which reduce them if justified. Another comenter stated that exact guidelines are required to provide an interpretation of i decrease in the eff ectiveness of a provision. Some comenters suggested that a change to the method or timeliness of management review of supplemental specification changes does not necessarily contribute to a decrease in the effectiveness of a provision.
Several comenters stated that the time limit for reporting changes to supplemental :
specifications was too restrictive and reccmnended extending it f rcen 3 day to 10-30 days. Many others recomended that the NRC Resident Inspector's role in the l
acceptance or rejection of change be clearly defined and that the NRC ensure sufficient comunication between Washington and Regions to assure uniform 1 plementation of the acceptance / rejection policy. One comenter recomended that licensees be allowed to implement proposed changes 30 days af ter the NRC is inf omed of them. Other comenters suggested that Sections 50.36(f)(6) and (7) which deal with changes and proposed changes to supplemental specifications that involve conflicts with technical specification in the 01., a decrease in the effectiveness of a provision or specification or an unreviewed safety question be deleted in f avor of 10 CFR Section 50.59.
l Two comenters suggested that an amendnent to the license under Section 50.36(g)(2) should only be at the initiative of the licensee.
In regards to the Conditions of License (Section 50.54), several comenters recoment'ed that licensees be allowed to submit any infomation in support of l reinstatement or modifications of proposed changes for those that were revoked and l that licensees be provided protection against arbitrary and discriminatory l revocation of changes by Regional Adninistrators and suggested that the licensee be given advance notice of revocation with an opportunity to contend the revocation.
I
D-1 APPENDIX D. ESTABLISHING CRITERIA Table 0-1 presents seven general steps required to establish criteria for any purpose. Not all of these steps will be fomalized in the establishment of most criteria one deals with, but they should be considered. In the case of the AIF Subcommittee's effort to develop criteria for the improvement of technical specifications bv clearly defining their scope, these steps have been forntalized
, and have been used.
Table D-1 Steps Required to Establish Criteria
- 1. Problem definition
- 2. Requirement definition
- 3. Establishment of expected results 4 Establishment of proposed criteria and bases
- 5. Testing of proposed criteria and bases
- 6. Documentation of problems with proposed criteria
- 7. Iterate and finalize criteria.
t I
Problem Definition The first concern to be addressed in the establishment of criteria is to define the i problem one is attempting to solve. Without this, the criteria established may l achieve something entirely different than what is actually intended. By proper
{ definition one may also find that criteria are needed for both the form and content I of the subject, in order to solve the problem. Such is the case with the present technical specifications: both their fom and content need to be addressed.
The problems addressed were:
- 1. How to change the form of the technical specifications to optimize their effectiveness as an operational tool; and
- 2. How to control the content of the technical specifications to limit their scope to items of highest safety significance.
i Our goal is safe and econanical nuclear plant operation. In 1984 the NRC processed about 1500 license atendments. This large number of requests distracts the
! regulators and the utilities from matters of real safety potential. Likewise, the 4
complexity of the present technical specifications confuses both the regulators and the regulated. What is needed is a document which will assist the operators in running the plant safely, rather than distracting them from those things of most tportance to safety. Clean, clear rules on the content of technical specifications are needed to assist the regulators and the regulated to ensure safety.
- Recuirement Definition
, There are restrictions, limits, or bounds 'on what one can hope to accomplish through the criteria to be established. These restrictions can be Constitutional, legal, regulatory, practical, political, financial, temporal or simply perceived,
- for example. The difficulty of the task defines how femally 1
D-2 limits are defined. Simple criteria may not require conscious recognition of any .
requirements, while complex tasks can be led astray by unrecognized or artificial requirements.
One has the legal requirement of the Atomic Energy Act to include special nuclear caterial in the technical specifications, as well as provisions to protect the health and safety of the public. This has been broadened through the regulatory
- requirements of 10CFR50.36 to include "such other things as the Comission may require". There is a political requirement in that the interests of many parties,
! . t:ithin the industry and without, must be satisfied. There is a financial I
requirement in that a complete study done from scratch could cost more than what is to be gained. It is more practical to use information that is presently
! available. There are also time requirements; for example, the TSIP report on j tethods to improve the technical specifications was completed October 1,1985.
Rather than undertake a complete study from scratch to establish a set of technical i specifications which demonstrably provides adequate protection to the health and I safety of the public, one can start from the fact that the NRC has determined that i the public health and safety are adequately protected if the consequences of l certain accidents meet established criteria. These established criteria relate =
1 primarily to the radiological consequences of severe accidents. The accident ,
analysis presented in Chapters 6 and 15 of each plant's Final Safety Analysis '
Report (FSAR) describes the results of these evaluations, and demonstrates that the
- criteria can be met if the plant is operated within certain assumed bounds. Thus 1
the validation of assumptions of the accident analysis is a logical requirement for 4
a set of technical specifications that are sufficient to protect the public health and safety. However, this approach alone can lead to a seemingly unbounded set of technical specifications. One must ask, "Is it necessary to validate all the assumptions of the accident analysis through the technical specifications?"
There are other means to validate many of the assumptions of the accident analysis, j oth:r than using the technical specifications. The question is, which of these other means should be used, and for which assumptions?
i Establishment of Expected Results j With the establishment of criteria, it matters very much where one wants to go.
- One of the requirements is that the end result be usable. One of the earlier i alternatives proposed, to a plant's adopting a set of technical specifications, was
! to use the plant's FSAR as its technical specifications. Evidently that is one i place the industry did not want to go; no plant has chosen that option. It would yield a maximum of complexity combined with a minimum of operational usefulness.
J The expected results of the early-1985 industry effort were twofold: .
I
! First, it was expected that the plant operators would be relieved of some of their
- burden of non-safety-related activities.
- Second, it was expected that the number of license changes to be processed by the NRO would be cut about in half due to the division of the technical specifications
} as proposed by the draf t rule (47FR 11369, 3/30/12).
Let us examine these expectations:
i s
I
0-3 First, it is expected that, af ter the division, the volune of technical (operating) specifications will be about half their present volume. Based on the result of a pilot study conducted by the AIF Working Group on Criteria Development, this ,
Cxpectation appears reasonably achievable. It is also expected that this division '
c:ill produce a proportionate reduction in license amendments. Such a proportionate i reduction presupposes that the items to be relocated to other documents are equally troublesome in the number of license changes they are responsible for. There is
, skepticsm that this is true. For example in a recent group of changes being processed for a reload at a C-E plant, only two out of eleven items would be Gxpected to be located in other documents. Another four items in this group were l ~ t: hat are temed cycle dependent variables. If cycle dependent variable, or their i' values, are removed from the technical specifications, reducing the licensing changes by half could be realized. Establishment of effective criteria will be r:alized by recognizing the results the chosen criteria, as well as other improvements, will produce and ensuring that these results are the ones desired.
Second, it is not clear that any appreciable benefit will accrue to the plant op:;rators unless something is done to the fom of the technical specifications, as well as to their content. While relocating items to other documents will concentrate the operators attention on the items of more immediate importance to .
safety lef t in the technical specifications, by itself, this does not reduce the volume of infomation and requirements the operator must deal with. Without an improvement in form as well as content, the proposed improvements could have less affect than is attainable.
One thing that can be done to reduce the volume of information the operator must deal with is to remove some things completely from the technical specifications.
Suggestions are discussed in Appendix G.
Establishment of Proposed Criteria and Bases Once one has an idea where one is going, a proposed draf t criteria can be i
cstablished. It should also be possible to set down sound bases for the criteria proposed. If adequate bases cannot be established, different criteria must be sought. Such was the case with using the operator's span of control as a criterion. This was one of the criteria in the Subconnittee's initial draf t, but tas subsequently rejected as having no demonstrable relationship to the protection
) of the public health and safety.
]
It is better to start with a less complete set of criteria and build on it as one l works with it than to expect perfection at the first attempt. Any complex set of i
4 criteria will require iteration, and this should be expected as part of the process.
i The bases for any proposed criteria should be clear. If they are not, one is not i .sure why he is doing whatever he is doing, and will inevitably misapply the criteria in some cases. In the case of the technical specifications, the ultimate concern is protection of the health and safety of the public. The public's health is protected by operating the plant within limits and under conditions which have been demonstrated through the accident analysis to minimize radiological releases due to nomal and anticipated operational evolutions. The public's safety is
! protected by operating the plant such that the probability and consequences of
- accidents are adequately controlled.
I
\
D.4 Testinc of Proposed Criteria and Bases A proposed criterion is tested by applying it against a sample population and examining the results. In the case of the draft criteria, this was done by applying a number of proposed criteria to two sections of each Nuclear Steam Supply Systems (NSSS) vendor's standard technical specifications.
The' results of the pilot study were as expected. Some of the proposed criteria tere more effective and/or easier to apply than others. About half of the tested specifications were found to be subjects for relocation. The important part of this test of the proposed criteria was the identification of the problems associated with the draf t criteria.
Documentation of Problems with Draf t Proposed Criteria During testing of any proposed criteria one uncovers unexpected problems. This is esp;cially true when someone not involved in the development of the proposed criteria is used to perfonn the test, and documents his or her rationale for specific decisions.
l Some of the problems with the draf t criteria of Aoril 19, 1985 that were found in *
, the pilot study to split the technical specifications are provided in Table D-2.
These are used as examples of what might be found from such a study, and are not intended as criticism of the study. They are an example of the application of the steps this Appendix advocates.
Table D-2 Problems with the Draft Criteria of April 19, 1935
- 1. Definitions of terms were not clear, particularly what is meant by design basis accident analysis.
- 2. Needs to address what the other means are of validation besides the technical specifications.
, 3. Need to address level of importance in the accident analysis.
- 4. Need to consider level of importance of support systems.
- 5. Need to consider imminent f ailure.
- 6. Need to consider the definition of operability and how it is applied in the technical specification:.
- 7. Use of the Critical Safety Functions (CSFs) as a criterion presented potential problems. CSFs involve the emergency procedures, which may invoke violations of the technical specifications. CSFs involve multiple success paths, without specificity as to which one should be the one validated by the technical specifications.
- 8. Use of the operator's span of control as a criterion, although appealing, was not particularly useful.
D-5 I
Through the documentation of such problems one can get to the core of the issue being dealt with. Such problems can lead to the rejection of some criteria, and the modification of others. The problems also point out where supporting information must be developed.
Interate and Finalize Criteria ,
From the above discussion, one can see the need for iteration. The draft criteria cf April 19, 1985 was the first round discussed with the NRC. For historical purposes, Figure D-1 shows the April 19, 1985 draft screening criteria. A second version of the screening criteria (May 16,1985) is shown in Figure D-2, and the third (June 27,1935) version 'of the screening criteria is shown in Figure D-3.
The final criteria will be discussed later.
Figure D-1 Draft Screening Criteria of April 19, 1995 Is the existing technical specification limitation applicable to a structure, system, component, or process variable that: ,
- a. Is under the direct control of the control room staff ?
- b. Validates those assumptions relied on in the design basis accident' analysis which are not validated by other means?
- c. Is necessary to fulfill the critical safety functions?
Figure 0-2 Draf t Screening Criteria of May 16, 1985 Is the existing technical specification limitation applicable to a structure, system, component, or process variable that is necessary to fulfill the critical safety functions and validates those assumptions relied upon in the design basis accident analysis?
This criterion can be reflected as three Questions:
- 1. Is the limitation required to validate an-assumption relied on in the accident analysis?
- 2. Is the assumption adequately validated by some other means (i.e., is it necessary)?
- 3. What is the level of importance to safety of the assumption being validated?
Figure D-3 Draf t Screening Criteria of June 27, 1985 Is the existing technical specification LSSS/LCO applicable to a structure, system, component, or process variable that:
- a. Is used to reduce the likelihood of a design basis accident by early detection of a significant abnomal condition, or
- b. Is part of the primary success path of a safety sequence analysis and j automatically actuates to mitigate a design basis accident, or
D-6
'c . Is an initial condition of a design basis accident analysis.
As can be seen from these three examples, the development of a suitable criteria set is an evolutionary and iterative process.
Historical Development of Criteria ,
In the past a number of different criteria for the content or division of the technical specifications have been proposed or discussed. It is of interest to examine certain of these.
First, there was the proposed use of the FSAR as a plant's technical aptcifications, as discussed briefly above. Under present regulations, this would have involved incorporating the whole FSAR into the plant's operating license.
This would have been difficult to maintain and change.
Second there are the criteria contained in the present version of 10CFR50.36.
These are the criteria which have resulted in the present set of technical sp:cifications. One of the problems with these criteria is the inclusion of the phrase "and such additional infonnation as the Comission finds appropriate". This ,
sometimes has been interpreted to give carte blanche to the NRC staff to include pet items. Actually this is a sympton of a larger problem: a mechanism is needed to ensure that comitments are not lost, without having to resort to including them in the operating license.
Third, when the NRC's proposed rule of March 30, 1982 came out, a criterion was developed based on the frequency of the surveillance requirement associated with each specification. If the surveillance was longer than a given period, say daily, the item was considered not to be of immediate importance to safety, and thus could be relocated. This criterion was not-totally accepted by the NRC staff.
Fourth, was a proposal from a group of seven utilities that met with the NRC to discuss the March 30, 1982 draf t rule. This proposal was to retain in the Tcchnical Specifications those 3rimary plant systems essential to achieve the four safety functions considered to se of imediate importance to safety: protecting the integrity of fission product barriers, controlling reactivity, cooling the fuel, and limiting the release of radioactive fission products following an accident.
These are the safety functions discussed in the General Principles section of the proposed rule. Specifically, this would include the three active safety function activities that are under the direct and constant control of the plant operations staff: values of process variables, operating state of equipment, and operating status of equipment.
Fif th, a number of form criteria were considered in early draf ts of the AIF cri teri a. These included: equipment or component listings, programatic requirements, surveillance requirements, cycle dependent variables, and bases as candidates for relocation. It was decided that these should be addressed elsewhere than in the criteria for division of the technical specifications.
D-7 Sixth,theoperator'sspanofcontrolwasconskderedasanattractivecriterion.
Unfortunately, no real basis could be found for this criterion. It did not relate to the protection of the public's health and safety, except in so far as the cp;rator has imediate and direct responsibility for the safe operation of the pl ant. It was also found that this criterion did not do anything that other criteria did not do, particularly the criterion of preserving the accident analysis assumptions. .
l Seventh, the critical safety functions received a lot of support as a criterion for l
division of the technical specifications, and perhaps, for the content of them.
- This criterion is more defensible than the others previously presented, although it must be linked with the accident analysis to establish those items necessary to establish the initial conditions for normal operation. This is an appealing criterion as much of the information is already available to support its application. One drawback of this criterion is that critical safety functions have
, betn used extensively to develop emergency procedures. This can result in an improper linkage between technical specifications and the emergency procedures of a
. plant, which may' call for violations of the technical specifications under accident conditions. The critical safety functions have an additional drawback in that there is usually more than one success path to accomplish each safety function. .
The use of a safety sequence analysis, as contained in the June 27, 1985 screening criteria, is needed to detennine which path is the primary success path. The primary success path is the one which assures a safe plant response, even without immediate operator intervention, as assumed in the accident analysis.
Eighth, the assumptions of the accident have been the criterion of the ANS 58.4 stcndard and the subject of previous work. Although these have a logical basis, they are believed to be more difficult to apply than some other criteria. There are problems associated with this criteria, which are not unique to it, but have been more closely identified with it. These include the question of other means of validation, and of level of importance to safety. Also, does this criterion mean the Chapters 6 and 15 accident analysis, or the whole FSAR? . As used in this section it means Chapters 6 and 15 only.
Ninth, a criterion was suggested based on the allowable outage time of a specification. The logic is that if the plant is not required to be shutdown within, say, twenty-four hours, the item cannot be of imediate importance to safety. An adequate basis for this criterion could not be established.
l Tenth, a risk-based criterion is possible. Due to the inherent problems with the acceptability of risk-based safety goals, it is' believed that such a criterion would be difficult to quantify at this time. Additional discussion on the use of l probabilistic methodology is found later in this section and in Appendix I.
Other criteria which have been suggested include:
- a. All systems and components which automatic actuation to prevent or mitigate an accident;
! b. Provide control of radiological releases; and
- c. Provide protection of the health and safety of the public (A very broad ranging criteri a).
Combinations of the above concepts have been considered. The suggested improvements include the fifth, sixth, seventh and eighth criteria in various l
combinations.. The end product of the industry effort are a set of recomendations -
(Section II of this report) which addresses both the fonn and content of the ttchnical specifications.
D-S Discussion of the Final Criteria In establishing a necessary set of criteria, it was first necessary to establish the intent of technical specifications. It was concluded that the intent is to specify the limiting conditions for operations to maintain the f acility in a configuration during nomal operation such that if an accident occurs, there is a high degree of assurance that the facility can be successfully brought to a safe shutdown state. These are automatic protection systems and components designed to correct the abnormal situations before a safety limit is exceeded. In some cases, operator action may be relied on as well as automatic actuation.
The regulatory basis for what limiting conditions should be included in technical specifications are those associated with the individual facilities design basis accidents. An analysis of a spectrum of design basis accidents by the staff is the principal means of perfoming a reactor safety evaluation. The " single f ailure criterion" concept used in the design of existing plants is preserved in this approach in that it is applied in the individual design basis accident analyses and is reviewed by the staff. The NRC's decision to issue an operating license is based partially on the conclusion that the consequence of the design basis ,
accidents are acceptable and that reasonable assurrance is thereby provided that the protection of the health and safety of the public is adequately maintained.
Considerable discussion was held on whether the design basis events should include natural events as well as transients. It was concluded that they should not.
Design basis transients are analyzed to assure fuel integrity and adequate system response capabilities. Natural events (floods, tornadoes, seismic events etc) are addressed to assure that equipment is not adversely affected by the event. Adequate procedural guidance is provided to ensure that the plant is placed in a safe condition given the occurence of a natural event. Additionally, equipment is qualified to survive such an event and still provide assurance that the primary success path is available.
Specific Criteria The three criterion selected and a discussion of the basis for each are provided.
Criterion 1: An installed system that is used to detect, by monitors in the control room, a significant abnomal degradation of the reactor coolant pressure boundary.
Discussion: A basic concept in the protection of the public health and safety is the prevention of accidents. Systems are installed to detect significant abnomal degradation of the reactor coolant pressure boundary so as to allow operator actions to either correct the condition or to shutdown the plant safely, thus reducing the likelihood of a loss of coolant accident.
This criterion is intended to ensure that technical specifications control those systems that detect excessive reactor coolant system leakage. Two specific examples of systems which are selected using Criterion 1 are:
Secondary system radiation monitors Reactor Building sisnp level instrumentation
D-9 Criterion 2: A process variable that is an initial condition of tne design basis accident analysis.
Discussion:
The intent of this criterion is to preserve in the technical specifications those initial assumptions in the design basis accident analyses for the specifi'c f acility that established the asstaned f acility configuration at the initiation of the design basis accident event. Process variables are parameters for which specific values
. cr ranges of values have been chosen as reference bounds in design basis accident analyses and which are monitored and controlled. Examples of these would include suppression pool water level, reactor coolant system pressure and temperature, condensate storage water tank . level etc. It would not include fuel characteristics set by the initial design and configuration of the fuel assembly.
As discussed above, as long as these variables are maintained within their technical specification limits, there is reasonable assurance that the structures, systems and components will perform as designed in protecting the health and safety of the public.
Criterion 3:
A structure, systed or component that is part of the primary success path of a saf ety sequence analysis and functions or acuates to mitigate a Design Basis Acci dent.
D.is cussi on:
This criteria includes in technical specifications those limiting. safety system settings and limiting conditioas for operation associated with structures, systems cnd components that represent the first line of defense and that functions or actuates to mitigate the accident. The f acility is designed with the necessary redundancy and diversity to address a multitude of design basis accidents.
Although there are several ways to respond to a particular accident (use of non-safety as well as safety-related designed systems), those associated with the pritary success path of. a safety sequence analysis should be included in the t chnical specifications.
A safety sequence analysis is a systematic examination of the actions required to nitigate the consequences of events considered in a plant's design basis accident analysis, R; as presented in Chapters 6 and 15 of the plant's Final Safety Analysis port. Such a safety sequence analysis considers all applicable events, whether explicitly or implicitly presented. The primary success path of a safety sequence analysis consists of those actions assumed in the design basis accident analysis which limit the consequences of the events to within the appropriate acceptance criteria.
Use of Probabilistic Methodolooies There was considerable discussion within the Subcommittee on why the criteria selected were deterministic rather than probabilistic methodologies. The criteria as defined do not in fact preclude future use of probabilistic approaches. In f act, as the probabilistic methodology matures it may be used to decide what-design basis accident analyses should be adeessed in Sections 6 and 15 of future FSARs.
Additionally, as discussed in other sections of the report and in the conclusions, the industry and NRC are encouraged to continue to develop thisatool and use it, where applicable ~, in the regulatory process.
D-10 As has been stated before, probabilistic analyses should support, not supplant, I de2enninistic requirements. Whatever their basis, regulatory requirements are l applied to the licensing of new f acilities or regulation of operating f acilities '
and must ultimately be reduced to a set of criteria, guidelines or directives which, ideally, should be (1) rational, (2) unambiguous to both regulator and applicant.or. licensee and (3) enforceable. While the great value of PRA Cethodology is in its rationality, it is complex in its implementation and interpretation. Therefore, while the current deterministic approach to setting requirements may be considered imperfect, it must be retained as the only practical approach. PRA should be used to make these deterministic requirements more
- rational, using plant operating experience and analytical approaches to identify additional design criteria which should be included or to identify areas where excessive conservatism should be eliminated Examples of improvements that have actually utilized this methodology to develop the necessary technical bases for the changes are changes to surveillance intervals and allowed outage times in technical specificati ons.
The Use of ANS 58.4(1979)
ANS 58.4 (1979) " Criteria for Technical Specifications for Nuclear Power Stations" is the nuclear corrunity's p:esent standard for both the content and format of
- technical specifications. It provides a conceptual framework for addressing the preparation of technical specifications. It contains criteria for selecting the
~
subjects and values to be included in technical specifications, as well as. criteria for developing the bases for the technical specifications. Such a standard is a necessary tool to collect the industry's thinking as to the philosophy behind the development and use of such an important part of a plant's operating license.
Revision to the industry standards take place every five years. It is appropriate that this should coincide with the other industry efforts now taking place.
Actually, the revision of ANS 58.4 -will lag the other industry efforts slightly, to take advantage of their experience, and to avoid.a . duplication of effort. Right now there are two criteria for the format and content of technical specifications:
the old 10CFR50.36 criteria, and the new criteria proposed by the draf t rule. Out of the present industry efforts will come a better definition of the criteria. The industry standard is an appropriate place to collect and explain this criteria. A writer's guide for technical specifications is also needed; it is planned to add this as an appendix to the standard. A revision is also needed to clean up sane minor problems which have been discovered through application of the.present standard.
r
E-1 App;ndix E: Examples of Technical Specifications Under the New System As a result of the pilot studies performed using the final criteria (see Appendix D), the following examples of structur~es, systems or components that may remain in the technical specifications are listed: -
Reactor Trip System Instrumentation
~
Engineered Safety Features Actuation System Instrunentation Radiation Monitoring for Plant Operations Reactor Coolant Loops and Coolant Circulation Safety Valves Leakage Detection Systems Reactor Coolant System Pressure / Temperature Limits ECCS Systems Containnent
- Integrity
- Leakage
- Air Locks .
- Air Temperature and Pressure
- Isolation Valves Main Steam Isolation Valves A.C. Electric Power Sources D.C. Electric Power Sources Some examples of structures, systems or components that would no longer remain in the technical specifications are:
Boration Systems (separate from those used as part df the ECCS injection function)
Seismic Monitoring Instrumentation Meteorlogicial Monitoring Instrumentation Remote Shutdown Monitoring Instrumentation Fire Detection Instrumentation Turbine Overspeed Protection Combustible Gas Control Penetration Roan Exhaust Air Cleanup System Vacuum Relief Valves Flood Protection Sealed Source Contanination Fire. Suppression Systems e
4
1 F-1 Appendix F: Disposition of Requirements not Appropriately Contained in Technical Specif1, cations The criteria outlined in Sections IV (B) and (C) can be applied to:
o Existing operating license technical specifications, custom or standard in origin; o Technical Specification under development for an NTOL plant; or o Proposed future changes to existing or new technical specifications.
Such applications will determine whether a given specification should remain (or be included) in technical specifications or be incorporated into other documents.
It is evident that af ter the criteria have been applied certain requirements will still exist that no longer will be controlled by the technical specifications.
Under this suggested procedure, these items would be incorporated into documents, each having a level of associated regulatory control. The ma,ior existing generic documents and associated controls include:
- o The FSAR: 100FR 50.59 currently allows a licensee to make changes to the facility or to procedures as described in the FSAR, or to conduct tests or experiments not described in the FSAR as long as no unreviewed safety question exists. Therefo e, for the former specifications which are appropriate to incorporate into the FSAR, a definite link between the actual procedures which control the item and the FSAR is proposed. TMs-link-wwld-conslit_of, a statement in the appropriate section of_the_ESAR-such-as, "The _ specific opeFaring intTts,_a_ssoci_atiq actionsmyd a test requirem_eots_wi1Lbe. as detailed in the appropriate f acility procedures".7 hen, if a change is proposed to one oT these proTedures, a safety evaluation would have to be performed and submitted as part of the required annual report to the NRC.
It should be noted that a similar FSAR statement will be provided for surveillance requirements: 1.e., tests associated with those LCO's which, per the cdteria, will remain in the technical specifications (See Section IV.B) of this report for discussion of relocation of surveillance requirements).
Although it is recommended that surveillance requirements no longer be included in the technical specifications, it is recognized that they are still a significant measure of the fulfillment of the OPERABILITY requirements of a given LCO. Therefore, the appropriate level of control in the following areas must be applied:
"3/4.0" requirements: these generic requirements will continue to be valid for surveillances. Therefore, their administrative application must be visible and tightly controlled. As a minimum, this will be done via surveillance procedures, each documenting these provisions (or by reference to a single administrative procedure which documents the provisions).
F-2 Interpretation of Operability: For ease of operator and NRC. inspector understanding, the surveillance procedures should clearly indicate (again, appropriate references are acceptable) how each requirement affects the operability requirements of the associated LCO. Each procedure should provide a clear reference to the LC0 it supports.
~
Surveillance Matrix: 0ne document should provide a clear reference for
.which procedures implement the surveillances supporting each LCO. This could be ref erenced in the FSAR.
o The QA plan: those items chosen to be incorporated into the QA Plan shall be controlled by the existing 10CFR 50.54a and 10CFR50, Appendix B.
O Inservice Inspection / Inservice Test Program: Those items chosen to be incorporated into ISI/IST Programs shall be controlled by the existing 10CFR 50.55a, (a) through (g) and 10CFR 50.59 if the items are also located in the F SA R.
o Fire Protection Programs: those items chosen to be incorporated into a Fire .
Protection Program shall be controlled by 10CFR 50.59 if the program is part of the FSAR. Otherwise, it is proposed that a cormiitment be provided by the licensee to perfom a " decrease in effectiveness" evaluation similar to that required for the physical security plan, o ODCM/PCP: those items chosen to be incorporated into either the Offsite Dose Calculation Manual or Process Control Program are proposed to be given the same level of review and control as currently provided via the RETS requirements.
The documents listed above will become the major sources for the relocated requi rements. If, however, a licensee justifies the use of alternate docurnents with sufficient control, this should be acceptable.
Finally, the locations of the bases for technical specifications should remain the licensees option, but in reviewing its options, the following should be considered:
- 1. Is the infomation correct and up to date?
- 2. Will the information be controlled as ~ changes are made?
- 3. Is the infomation located in a manner that will allow ease of use by those needing to make interpretations (i.e., plant operators, supervisors, licensing, NRR,I&E,etc.)
To more fully explain the suggested process for those items no longer located in the technical specifications, Figure F-1 provides an example. The following steps would be taken:
The criteria would be applied to existing, new or suggested technical specifications; For those items not meeting the criteria, they would be located in other documents (FSAR, QA Plan, ISI/IST Plan etc., as discussed above);
F-3 For those topics located in the FSAR, a statement would be made in the
_ appropriate section of the FSAR that, "The specific operating limits, associated actions and test requirements will be as detailed in the appropriate f acility procedures";
The detailed requirements (Operating limits, Action Statement, Reporting Requirements, Applicable modes, etc) would be placed in the appropriati procedures, whether operating, test, maintenance, surveillance, etc. exactly as they are presently found in the technical specifications; and The regulatory control suggested is the present 10CFR 50.59. It allows the holder of an operating license to "make changes in the procedures as described in the safety analysis report" if it does not involve an unreviewed safety question.
The use of this regulatory standard to control changes to those items no longer required in the technical specification is of equivalent control in that:
- It requires a detailed review to determine if (1) the probability of occurrence
- or consequences of an accident previously evaluated may be increased, (2) the possibility of a new or different kind of accident may be created and (3) the margin of safety is reduced; It is auditable by I&E in that the records of the evaluations of the chances ~
must be maintained by the licensee (10CFR 50.59 (b));
- The licensee must furnish to the NRC Regional Office with a copy to the Director of Inspection and Enforcement, a report on the changes at least annually or at shorter intervals as may be specified in the license (10CFR 50.59 (b));
The licensee must update, pursuant to 10 CFR 50.71, the FSAR annually in addition to providing a brief description of changes made under 10CFR 50.59 For those LSSS/LC0's remaining in the technical specifications, the associated surveillance requirements would be located either in surveillance procedures, a surveillance plan or a specific licensee program addressing surveillance. It is recommended that only these be labeled "surveillances".
Those associated with items no longer located in technical specifications should be labeled " tests" or some other appropriate title. This would distinguish those checks associated with technical specification requirements from the other requirements.
The location of the Administrative Sections of Chapter 6 of the technical specifications describing the f acilities' staff qualifications and both the Fac'lity Safety Review and Corporate Nuclear Review and Audit Groups could be relocated in a licensee doctanent such as the QA plan, FSAR, etc.
F-4 Conclusion -
Having reviewed the existing generic documents available for those topics presently in the technical specifications that would be located elsewhere, there are adequate documents and existing regulations controlling those documents. There is no need to establish a new document entitled " supplemental specifications". This would unnecessarily add.to the already overburdened adninistrative process that both the licensee and NRC staff have to use. There is no need to establishing additional
' regulatory controls other than those that presently exist. These include,'but are not limited to,10CFR 50.54a, 50.59 and 50.55 (a) thru (g).
It is the position of this Subcommittee that the process outlined above will provide an appropriate level of control of those items which will be removed from the technical specifications via the criteria. The present regulatory controls Dill not-degrade the current level of safety of plant operations.
l 1
e em w
. . - + - ,, , _ . , , _ _ -. -- _ m..- -- . _ _ _ _ _ . .v _. , ,y _ ,_. m,._ _y__ , . . , _ _._,,--,,.-.-,,,e. . , , , . -- _ _ - . - ~ - ,.,v _ - . , . , ,
Existing /New/ Proposed '
Technical Specification Figure F 1
- Surveilknee Surveillance :
procedures I
o
- Apply cdcria
/ \,
Other t Revised documents 4
technical specifications -
! n ,
.. Astatorentsetne FSAR "i' 6* k5 "6" '
Consists of: ""
1 "'d ,g'"l'*/i, C'"" '
. Definitions praedura.
- s
. Limiting safety system settings
- . Limiting conditions for operation
. Action statements
. Applicable modes i
. Specialmaterial
. Specialtests Operating procedures l
Will not require:
. Surveillance requirements n Test / maintenance
. Administrative details procedures
. Design information Where applicable,
. Component lists detailed:
Other
. Operating limits
- . Action statements w procedures
. Operating mode
. Test requirements l
I i
- Alternative generic decament I
- oA plan (App. 3)
-isvisT (1scFR50.55)
- Fire protecties (10CFR50.58)
Note: There may be acceptable specific - occM (RETs progree) licensee alternative decaments- - PCP (RETs progress) l
G-1 Appendix G: Discussion on Short Tem Solutions Not Requiring Rulmaking There are several significant improvements in technical specifications that can be realized without requiring rulmaking. These improvements can go forward now by, for example, the issuance of generic letters as was done with snubbers. The following six short term improvements and suggested actions are discussed in this Appandix:
~
Federal Regulations duplicated in the technical specifications;
- - The definition of Operable-Operability;
- The 18 month surveillance intervals; Cycle specific variables; and
- Applying Specification 3.0.4 consistently.
Removing Lists of Components The short term solutions are not limited to just these six and this list should be augmented as appropriate.
- 1. Federal Regulations duplicated in the Technical Specifications.
The proble to be addressed is the need to place duplicate requirements in the technical specifications that are found in the regulations. It is recomended that these requirements be deleted fran the technical specifications.
In cases where the requirement needs to remain in. technical specifications for clarity, the requirement would be replaced with a reference to the regulations or the program that implements the requirement.
The primary basis and justification for this recomendation is that the following statement taken from operating licenses issued by the NRC indicates that licensees are required to comply with federal regulation as well as plant technical specificationr.
"This license shall be deemed to contain... and is subject to all applicable provisions of the Act and to the rules, regulations... and is subject to the additional conditions specified or incorporated below.
(2) Technical Specifications...."
Ai; a result of this license condition, there is no need to repeat federal regulations in technical specifications to ensure compliance with these requirements. The duplication of requirements only adds to the administrative burden of both the NRC and licensee.
Additionally, NRC Generic 1etters 82-17 and 82-23 indicated that licensees are required to meet the requirements in regulations regardless of provisions in technical specifications prior to the issuance of the regulation or the provisions that have been added to technical specifications since the regulations were issued.
The implementation of this recomendation will not relieve the licensee of the requirement to comply with any current applicable regulations or technical specifications. As a result, the change will have no safety significance but will relieve the administrative burden of the NRC and the licensee. This will allow both the NRC and the licensee to redirect manpcuar resources to other areas of
. plant operations.
I G-2 i l
Suggested Regulatory Action to be Taken j An NRC Generic Letter should be prepared for issuance. This letter would provide guidance to licensees for the deletion of requirements from technical sp;cifications that are also present in federal regulations. It would also request that licensees submit technical specification changes which incorporate this guidance. -
- 2. The Definition of OPERABLE-OPERABILITY The current definition of OPERABLE-OPERABILITY in the technical specifications is ambiguous. This ambiguity leads, in many cases, to overly conservative application cf the definition, and unnecessarily declares equipment or components inoperable.
In addition, there is no well-defined regulatory basis for the current definition cf OPERABLE. This problem is exacerbated by the provisions of technical specification 3.0.3. and 4.0.2, which imply if you cannot positively assure that Equipment is OPERABLE, inoperability must be assumed.
The problems with the current application of the definition of OPERABILITY stem from a conservative application of the definition to cases which do not warrant .
such application. The definition in itself does not recognize conditions where it is still possible to demonstrate the equipment is capable of performing its design function, even though all atinistrative conditions may not be fulfilled "to the l et ter."
Clearly, the design basis includes application of all accidents and transients analyzed in the FSAR, but does not include simultaneous application of those accidents with a SSE-(Safe Shutdown Earthquake) or DBE (Design Basis Earthquake).
This is reinforced by the NRC in a recently draf ted internal position statement regarding " Technical Specification Operability Requirements." Technical Specification LCO's are designed to assure "those systems, subsystems, trains, components or devices which directly provide the capability to mitigate the .
consequences of design basis events, as well as generally specifying LCO per those which indirectly provide this capability as a support function."
In the Environmental Qualification Rule,10 CFR 50.49, (48FR2729) the NRC published certain criteria to demonstrate that near term operating license plants (NT0Ls) could continue operation in spite of not fully meeting the requirements of the rule. This proposed change applies those same standards to all decisions regarding Equipment qualification problems (both seismic and environmental). This analysis would demonstrate acceptable determination of OPERABILITY, by showing that the squipment is capable of perfoming its design function, or that alternative cquipment can satisfy the safety function.
Regarding missed surveillances, the current ' application of the OPERABILITY definition, as well as Standard Technical Specification 4.0.3, leads to unnecessarily declaring equipment inoperable. This concept implies " guilty until proven innocent" philosphy, and subjects the plants to unnecessary transients, shutdowns, and improper application of valuable resources. If a surveillance interval is missed for a piece of equipment, the surveillance should be perfomed as soon as possible. Should the equipment f ail to satisfy its surveillance requirements, it should be declared inoperable. However, declaring equipment inoperable due to an atinistrative oversight without regard for the actual equipment condition is unnecessarily conservative.
G-3 All of these changes clearly f all within the intent of the definition of OPERABILITY, which is to ensure that " equipment is capable of performing its spIcified function when all necessary instrumentation, controls, normal and emergency electrical power sources, cooling, or seal water, lubrication, or other auxillary equipnent that are required for the system to perfom its function are also capable of performing their related support functions." .
SUGGESTED REGULATORY ACTION TO BE TAKEN Reconsnend suggested changes be issued via Generic Letter, saying "the NRC staff has determined these clarifications are acceptable and licensees may submit proposed changes for NRC approval at their option." These should be incorporated into all 4 NSSS Vendor Standard Technical Specifications. The following additional language that is underlined is suggested.
, 3/4 LIMITING CONDITIONS FOR OPERATION AND SURVEILLANCE REQUIREMENTS 3/4.0 APPLICABILITY LIMITING CONDITION FOR OPERATION .
3.0.5 When a system, subsystem, train, component, or device is determined to be inoperable solely because a supporting system, subsystem, train, component or
, device is inoperable and a Limiting Condition for Operation and ACTION requirement (s) for that support equipment are specified in the Technical Specifications, operation may continue consistent with the ACTION requirement (s) i for the LCO for that support equipment.
APPLICABILITY SURVEILLANCE REQUIREMENTS 4
i 4.0.1 Surveillance Requirements shall be met during the OPERATIONAL MODES or other conditions specified for individual Limiting Conditions for Operation unless otherwise stated in an individual Surveillance Requirement.
4.0.2 Each Surveillance Requirement shall be performed within the specified time interval with:
- a. A maximum allowable extension not to exceed 25% of the surveillance interval, but
- b. The combined time interval for any 3 consecutive surveillance intervals shall not exceed 3.25 times the specified surveillance interval.
4.0.3 Failure to perform a Surveillance Requirement within the specified time interval shall constitute a failure to meet the OPERABILITY requirements for a Limiting Condition for Operation from the time of discovery. The Surveillance Requirements shall be completed within the time limits of the associated ACTION requirement (s) if plant conditions permit. Those Surveillance Requirements that cannot be rompleted due to plant conditions must be completed at the earliest opportunity when plant conditions permit. Exceptions to these requirements are stated in the Individual Specifications. Surveillance Requirements do not_ have to b? performed on inoperable equipment,
,r y---w- w- - - -my, -- - - - --- --- ,*rw ,*r - -
-t- ^ --- --c---? N r-*+ + - - - W-I---* *--
G-4 OPERABLE - OPERABILITY l.6. A system, subsystem, train, component 'or device shall be OPERABLE or have OPERABILITY when it is capable of perfoming its specified function (s), and when all necessary attendant instrumentation, controls, normal and emergency electrical power sources, cooling or seal water, lubrication or other auxiliary equip' ment that are required for the system, subsystem, train, component or device to perform its function (s) are also capable of perfoming their related support function (s).
- Under certain conditions where a component or device is not fully OPERABLE. the associated system, subsystem or train may be demonstrated OPERABLE or to have OPERABILITY, so long as an analysis is satisf actorily performed. Under these circumstances, the licensee may perform an analysis to justify continued ODer ati on. This analysis can include, where appropriate, consideration of:
- 1. Accomplishing the specified safety function by some designated alternative component or device if the principal component or device has not been demonstrated to be fully OPERABLE.
- 2. No significant degradation of any safety function or misleading
- information to the operator as a result of f ailure of a component or device resulting f rom a design basis event.
- 3. Limited use of administrative controls over a component or device that has not been demonstrated to be fully OPERABLE.
- 4. Completion of the safety function prior to exposure to the accident environment resulting f rom a design basis event and ensuring that the subsequent f ailure of the component or device does not degrade any safety function or mislead the operator, if applicable.
The validity of partial test data in support of the original
~
5.
qualification, if applicable.
3- The 18 Month Surveillance Interval With the advent of longer fuel cycles and less frequent and longer outages, the nuclear industry is encountering difficulty meeting the 18 month surveillance interval specified by Technical Specifications. As more
- utilities go to longer cycles this problem will intensify.
Initially the average fuel cycle lasted 12 to 18 months dependent on plant perf om ance. Outages, which before served primarily to allow refueling and surveillance outages, are now predominantly to make modifications. Refueling and surveillance outages are typically five to six weeks long while modification outages are generally three to four times longer. These outages have become this long primarily due to the extensive modifications mandated by new regulations and requirements (NUREG-0737, Equipment Qualification, Appendix R, Regulatory Guide 1.97). Consequently, over the past several years utilities have gone to 18-24 month cycles to improve power availability, fuel t utilization and to reduce shutdowns. Thus, where utilities had a total interval of 12 to 18 months between shutdown tests, the current trend is toward an interval of 18 to 24-months.
Over one hundred of the Standard Technical Specifications (STS) surveillances must be perfomed at 18 month intervals. Specification 4.0.2 allows this interval to be extended by as much as 25%. Specification 4.0.3 allows an even longer interval between tests provided that the equipment is not required to w - , - ---g . - - , - - - - - , - . - - - - -.
- - . -,--r e-r--- , --.~
G-5 be operable. For a single interval these allowances generally provide sufficient flexibility to reach the next planned outage. Over time however, the limit, not to exceed 3.25 times the stated interval for any three consecutive intervals, beccnes impossible to meet. For ex mple, recent reviews of the upcoming cycle indicate that a B&W plant will exceed the 1.25 limit for six surveillances and that over thirty surveillances will exceed the 3.25 limit. If a forced shutdown does not occur during mid-cycle, either a surveillance outage or massive technical specification relief will be necessary. In many cases the 18 month surveillance test must be done in Modes 4 and 5 resulting in making forced outages longer
~ without proper preplanning. Thus efficient operation of the unit is punished by a shutdown solely to comply with 18 month intervals. Neither the additional shutdown and associated transient nor the " band aid" solution of one time Technical Specification relief are desirable The 18 month surveillance interval is based on what was perceived to be the expected fuel cycle length. This surveillance interval was established during the original development of STS af ter discussions with senior NRC staff members and reactor and fuel vendors. A maximum time period of 22.5 months was thought to be sufficient to accomodate scheduling and performance considerations. 'A nominal .
interval of 24 months allows a maximin interval of 30 months which is considered unacceptably long by the staff. The staff has not supplied a technical basis for this unacceptability.
It is recornended that, because the 18 month surveillance interval applies to so many surveillances, several options should be pursued to alleviate the problem.
, These options are:
- 1. Delete the 3.25 criteria for all 18 month surveillances.
- 2. Delete "during shutdown" from those tests which can be safely and efficiently performed during operation.
- 3. Change some surveillance intervals to 20 to 24 months.
, The basis for this reconmendation is that in general, when evaluating a technical specification change, an evaluation of the revised specification pursuant to its t
j
~
initial basis is sufficient to assure an adequate level of protection. However, the current trend by the staff appears to be the application of an arbitrary, undefined level of reliability as a basis for technical specification frequencies.
This trend is illustrated in the March 30, 1982 draft rule Section (f) (3):
A change to monitoring provision is deemed to involve a decrease in the effectiveness of the provision if: (1) the frequency of the monitoring inspection, testing or calibration is decreased without a compensating change in the acceptance criterion or an increase in the sensitivity of accuracy or the ~ method used, unless the cumulative history of the test results clearly supports a reduction in frequency.
As a result, it appears that any extension of the surveillance interval must address compensatory actions or reliability.
There are several problems with the use of the draf t rule's criteria. One problem is that all equipment is treated equivalently regardless of its importance. This leads to the imposition of overly conservative requirements. It also assumes that
i G-6 the current interval is not overly conservative even for the most important . i equipment. An additional problem is that acceptance criteria in several specifications is already as tight as possible. Although not specifically required by the technical specifications, it is ccmnon practice to take "as found" and "as lef t" data before and af ter test adjustment. Typically the "as lef t" tolerance allowed is much more stringent than the "as found" acceptance criteria. This practice helps to assure that equipment found close to the allowable limit is not allowed to go through the next cycle without recalibration. Finally, efforts to
~
establish a ctraulative history of a piece of equipment are generally very involved and tedious. Especially so, since testing practices may have changed over the 4
history of the equipment.
Deletion of the 3.25 Criteria:
Deletion of the requirement "any three consecutive intervals must not exceed 3.25 times the interval" will not significantly effect equipment reliability. The current criteria allows a 22.5 month interval for as many as two intervals daring a three interval period. Deletion of the 3.25 criteria will allow all three intervals to be 22.5 months long. Per specification 4.0.2.a the staff has already ,
accepted that a 22.5 month interval will provide a sufficient level of protection.
Allowing this interval to be applied to all cycles will maintain a constant level of protection.
Deletion of "During Shutdown"
, The staff has already accepted this option, at least in part, since the Standard Technical Specifications allow performance of several 18 month Surveillances dJPing operation. Additional tests which are currently required to be performed "During Shutdown" by STS, but which could be safely perfonned during operation should be addressed individually. A review and evaluation of the STS to determine if additional specifications should be revised to delete "During Shutdown" could be i performed by Owners Groups.
Extension of Specific Intervals:
]
There are several reasons that interval extension should be allowed. These are:
- 1) Some equipment should not be significantly affected by the addition of four to 2
six months to the interval. For example: the visual inspection of the contaiment sump for debris or structural distress would generally not indicate significant additional degradation due to extending the surveillance i nterval .
- 2) Enough operability tests and checks are performed during operation such that extension of some intervals should not be a concern.
- 3) Scne equipment is not important enough to warrant an 13 month interval surveillance. For example, post accident instrumentation does not perform any automatic safety function nor is accuracy a major concern with these-i ns trtrnents.
SUGGESTED REGULATORY ACTION J The staff should issue a Generic Letter allowing the implementation of the I
G-7 recomended options. Implementation of these options should not generically necessitate the utilities' acceptance of additional STS requirements. The Generic Letter should clarify the acceptance criteria that must be applied to allow lengthening of surveillance intervals. This criteria should be sufficiently flexible such that a detailed analysis or extensive document search is not required in every case. The resultant technical specification changes would be permanent.
- 4. Cycle Specific V giables_
. Cycle specific variables are presently included in technical specifications. The
! basis for this is found in 10CFR 50. 36(b) as well as 10CFR 50.36, Sections (ii)
(A) and (2). It is recormended that these variables be relocated to another document, specifically the FSAR, and controlled by 10CFR50.59.
The NRC and industry are aware that a significant number of requests for license amendnents involve .a change of those variables that still clearly remain within the acceptance criteria previously approved by NRC, but change only because of a fuel reload for a later cycle. Tnis f act is reinforced in the " Current Problen" discussion published by the NRC in a proposed change to 100FR Part 50, Technical Specifications f or Nuclear Power Reactors, 47 FR '13369 dated March 30, 1932: -
"Moreover, the increased voltrie and detail of Technical Specifications and the resultant increase in the . number of proposed change requests that must be proposed have increased the paperwork burden fo* both licensees and the NR:
, staff."
The industry documented the same problem in a letter fron J.J. Shephard, Chairman, Westinghouse Owners Group, to J.H. Sniezek, NRC, dated October 4,1933.
Under current regulations,10CFR 50.59 permits licensees to evaluate new core reloads without prior NRC approval, unless tFe reload involves an unreviewed safety question or a change in the technical specifications.
! Industry and NRC experience have demonstrated that core reloads that do not involve an unreviewed safety question also do not represent a significant hazards
]
consideration as defined in 10CFR 50.92.
Therefore, the need for public scrutiny in these instances is obviated, and the
, application of the prenoticing requirement becomes little more than paperwork shuffle and unnecessary delay in issuing the reload license. In some instances, the date of issuing the reload has been critically close to the facility restart date, thus presenting the risk of delaying the facility's restart. Although this 1 has not yet occurred, the large number of reload license requests could eventually result in a case where the NRC staff f aces a choice to either delay restaat or forego the public notice process.
1 Since the 10CFR part 50.59 process has been in effect since December 1953, the industry and NRC have had substantial experience with application of these criteria to proposed changes to the facility. This process is already applicable to core reloads which do not involve a change to the technical specifications. NR:
approval would be required if the reload were determined to involve an unreviewed
! safety question.
i
l G-8 E
SUGGESTED _REGULATORYACTIONTOBETAJ_EN NRC publish a " position letter" stating that cycle specific parameters may be i
relocated to the FSAR and changed via 10:FR 50.59. The LCOs should remain and the ACTION, MODE APPLICABILITY, and SURVEILLANCE requir6nents should not be changed.
- 5. A,pp1ying Specification 3.0.4. Consistent 1y' Exceptions to the provisions of Specification 3.0.4. are not consistently applied I throughout the technical specifications. This lack of flexibility restricts plant
! startup for no apparent reason.
! As stated in the Standard Technical Specifications:
1 ,
1 3.0.4 This specification provides that entry into an OPERATIONAL CONDITION must be made with (a) the full conplenent of required systems, equipment or f
components OPERABLE and (b) all other parameters as specified in the Limiting .
- Conditions for Operation being met without regard for allowable deviations and out of service provisions contained in the ACTION statements.
- The intent of this provision is to enstre that unit operation is not initiated 1 'with either required equipment or systems inoperable or other limits being i exceeded.
1 i Exceptions to this provision have been provided for a limited number of 4
specifications when startup with inoperable equipment would not affect plant safety. These exceptions are stated in the ACTION statements of the
- appropriate specifications."
? It is recomended that the NRC place definitive generic criteria for 3.0.4 ;
j exceptions in the bases, review the Standard Technical Specifications and 1 ensure consistent application of the criteria, i
! As the basis for this recomendation, Specification 3.0.4 states:
3.0.4' Entry into an OPERATIONAL CONDITION or other specified condition shall not be made unless the conditions for the Limiting Condition for Operation are i met without reliance on provisions contained in the ACTION requirements. This provision shall not prevent passage through or to OPERATIONAL CONDITIONS as 1 required to conply with ACTION requirements. Exceptions to these requirements
- are stated in the individual Specifications."
1 I
As defined in Bases Section 3.0.4, " exceptions.... have been provided... when startup with inoperable equipment would not affect plant safety." Therefore, if it
- can be shown that this general criteria is met, an exception is appropriate.
J f
1 i
l G-9 l l
A review of the Standard Tech Specs results in the following more definitve l criteria for applying an exception to 3.0.4: '
- 1. If an alternate means is provided within an action statement to achieve the LCO, and unrestricted continued operation is then permitted. Ex mples are:
o isolating valves to achieve containment integrity o placing instrument channels in a tripped condition to ensure their safety function is met o instituting a " preplanned alternate method"
- 2. If continued operation is allowed based on the submittal of a Special Report.
- 3. If, under one specification, an action requires an affected system that is governed by a different specification to be declared inoperable, a 3.0.4 exception should be allowed if it is provided in that different specification, regardless of the lack of a 3.0.4 exception in the originating action. .
These criteria for judging the appropriateness of an exception to 3.0.4 are based on exceptions currently allowed by Standard Technical Specifications: unfortunately they have not been applied consistently.
Suggested Regulatory Action'To Be Taken The above criteria should be advocated via an NRC Generic Letter. The Generic .
Letter should encourage licensees to do the following:
- 1. Add the criteria to the Bases for 3.0.4
- 2. Review their technical specifications against the criteria and propose changes to achieve consistent application.
- 6. Removing Lists of Components Standard Technical Specifications currently contain :avaral different component listings, such as:
- 1. Fire Detection
- 2. RCS Pressure Isolation Valves
- 3. Contairinent Isolation Valves / Dampers
- 4. Spray / Sprinkler Systems
- 5. CO2 Systems
- 6. Halon Systems
- 7. Fire Hose Stations
- 8. Containment Penetration Overcurrent Protection
. _ =. - - - . _ . . . - _- ._ _
G-10
- 9. MOV Thermal Overload Protection
- 10. Steam Line Safety Valves
- 11. Secondary Containment Bypass Leakage Paths ,
l The existence of these listings of equipment numbers creates the need for license amendments to change a number, to add a component, or to delete a component
- regardless of the safety significance. Some of these amendnents have to be treated as " emergency" changes because the level of detail in the listing has legally created non-compliance with an LCO when no safety concern exists.
. The following excerpts are taken from Generic Letter 84-13 " Technical l Specification for snubbers" (May 3,1934):
, o "During the last several years, a large number of license amendnents have been required to add, delete or modify the snubber listing within the technical specifications. We have reassessed the inclusion of snubber listings within
- the technical specifications and conclude that such listings are not necessary provided the snubber technical specification is modified to specify which snubbers are required to be operable."
o Since any changes in snubber quantities, types, or locations would be a change to the f acility, such changes would be subject to the provisions of 10 CFR Part 50.59 and, or course, these changes would have to be reflected in the
- records required by paragraph 4.7.9.f." .
- In the enclosure to the generic letter, the LCO for snubbers is altered to be more 3 explicit as to which snubbers were required to be operable, apparently based on the deletion of the tables.
Based upon the above precedent, the NRC position appears to be as follows:
- 1. Listings of components enforced by an LCO cause unnecessary problems.
- 2. Changes to the components affected by an LC0 are governed by the requirements of 10CFR50.59
- 3. Licensees must keep documented records associated with such changes; this is currently covered by the following item under " Record Retention" in Section j 6.0:
The following records shall be retained for the duration of the Unit Operating License:
... Records of review performed for changes made to procedures or equipment or review of tests and experiments pursuant to 10CFR50.59 This position is easily applied to each of the listings provided above.
i i
= .
G-11 S u g g e s t ed je gul,a t,ory, ,A c,t 1,09 ,to ,b e, ,T,a,k,e n, It is suggested that the NR provide a generic letter si.nilar to 84-13 that woald allow deletion of the aforenentioned listings and any other approximately justified plant-specific listings. The letter would:
- 1. Direct each licensee to incorporate the current listings into the surveillance procedures which are associate 1 with the a'fected egalpnent,
- 2. Either propose or request proposal by the licensee of any necessary LCO clarificatioqs based upon deletion o' the listings, and
- 3. Allow deletion of the listings (and all reference thereto) based upon a proposed amendnent that acknowledges 1 and 2 above.
It is the industry positioq that any enforceability concerns would be rendered moot by the above requirements.
i a
9
H-1 Appendix H: Suggested Changes to the Regulations No 1cw or rule precludes the use of the suggested criteria discussed in this report. The Atomic Energy Act and rules do require Technical Specifications and do sp::cify fo-mat and derivation ("from the safety analysis"). They do not specify the scope of Technical Specifications nor the proper level of detail.
Since 1974, the scope of the " Standard" Technical Specification has greatly expanded. This is reflected in the almost 50 percent ir. crease in doctinent voltne
. for many licensees. This increased bulk is due to lack of firm criteria delineating scope and level of detail appropriate for technical specifications.
4 Tho AIF Subconittee and the NRC-TSIP have had discussion on the criteria and have now ccine to an agreecuent on what the criteria should include. That understanding 4
is spelled out in the three criteria. discussed in Section IV and Appendix D of this report. Just as technical specifications have grown enormously in the last decade without rule change, so too could they be cut back without rule change.
However, the same forces that caused growth over the last 10 years would rensin.
lI Without formalization of the scope of technical specifications, the documents will =
restrne their inflationary growth.
The industry believes that the fornalization of scope should be arrangej through inclusion in 100FR 50.35 of the three criteria. Other processes (Policy, j Regulatory Guidance, NR" '4anual changes) are subject to unilate al modifications j with limited industry or public participation. Rulenaking allows industry and
, public participation in both the initial rule and, more importantly in this case, in any change to the rule. -
This fonnalization of the change process yields confidence in the long-term 3 stability of the boundaries within which the industry must operate. It is this stability which is needed for the long-term improvements in technical specifications.
The following changes to 107R 50.36 as underlined are suggested.
proposed changes to 10CFR 50.35 (a) Each applicant for a license authorizing operation of a production or utilization f acility shall include in its application proposed technical specifications in accordance with the requirenents of this section. The technical specifications must be (Srived from the anal,y,ses and evaluaTf5n
~
includidTri t6e'Tifety ana[yjfs report and'iniiddiiEdis thereto iutinifffid~under
^ ~
50.36. These specificltidns Ei dis ~dibid~in paragraphs l t h i s s ec t Idri.~~ ,T 68 ~b~is e s oF~EEas 6ris (di
~
~f 66iTs6 mustle ~ike,c ff f diffon s(d[ ~!
~
included Tn th i specificaffon,s,e,_a_gpMfog but wfff ~riof ~bddome ,part o(Thi~ffi6rifddf"
! (b) Each license authorizing operation of a production or utilization' facility of 4
a type described in 50.21 or 50.22 of this part will include technical specifications. For a nuclear reactor operating licensing issued before (183
~
days af ter the effecifve~date of t1iis ani6dnT6t) and for a fuel re,grocessing pTi6f thi TfcE6si~ifff~f6EftIde tic ~finTeiT~ specifications in the categories set (6F#1i f6 paragrip] ]~[~df ~f6fs fifffon, for a nuclear reactnr operating
~
} ffcense issuid on or aftiFTfg5 days af ter the effective date of this iiiin~diii6fJT6ETffidii~wfff include filhdfiif specifications in the categories
~
IEf76ftTin paragrap"EI'(dT 6f~f6fs section. The Ccmnissioners may include Eddfff66 attic,h6fcif 7[eciffiatf66s~is they find appropriate.
_ _ - - _- - - ._ - - _ - - - - __ _ = - . - _ _ _ - - _ _ _ _ _ _ = . - _ _ - . _ _
e .
3 H-2 ;
(c) _ Technical specifications for a nuclear reactor operating license issued before ~ '
(180 difs after tfii efficTfve-~diff'6f This"did6disdf)~'indTdf'iTtIET"""~"~
- f66f6cdisTd{ pflif JfT1~fn?ftidiTfdis=1n thi76Tfoilng categories
- l i
j .....
1 i (d) Technical specifications for a nuclear reactor operating license issued on or .
I
, affe7TT66 diys-~ifter the effEctfvE date of this amendnent) wifl include in
{, agptqpttat,e detaX6cTy ffisse:
I 1. Installed systems that are used to detect, by monitors in the control I room, a,s,ffdfffdiff ibnornaf d,5cMi,on of the reactor coolant pressure l [gu,njr,yl,qr
, 2. Process variables that are a1 initial condition of the design basis
~~
j accfdi6f~iiiT,y'_ ifs; o-
- 3. Structures, Mstens tconponents that ar_e part of the primary success path of 'i~s~ifit~~~seSIEdie ~EdifNfs ~E13Tuncti01s or aftEtes to mftfriate a Bjfj5TiiYs accfdi5t.
- I The function of the technical spe:ifications is to maintain the f a:llity in su:h a configuration as to maximize a successful and safe shutdown and preserve the ,
validity of the design bases of the f a:llity. They address stru:tures, systems, conponents and process va iables needed to support the primary success path to
! bring the f a:llity to a safe shutio49. They are designed to control reactivity, s cool the fuel, prote:t the integrity of the fission product barriers and limit the release of radioactive fission products irmediately following an off-normal event. I Technical specifications 'are to be imposed on all normal modes of f acility
- operation including shutdown and refueling and are to consist of itens of the j following types:
! (i) S_af e,ty _limi ts Safety limits fo* nuclear reactors are limits upon important process variables which are found to be necessary to reasonable protect the )
- integrity of certain of the physical barriers which guard against the j uncontrolled release of radioactivity. If any safety limit is exceeded t5e
- reactor must be shut down. The licensee shall notify the Cormnission, review ;
the matter, and record the results of the review including the cause of the condition and the basis for corrective action taken to preclude recurrence, i Operation may not be resumed until authorized by the Cormnission (ii)LimitinLsafetysystensetting Limiting safety systen settings for nuclear i
reactors are settings for automatic protective devices related to those variables that meet one of the categories set forth in paragraph (c) of this section. Where a limiting safety system setting is specified for a variablo
!' on which a safety limit has been placed, the setting must'be chosen so that automatic protective action will correct the abnormal situation before a safety limit is exceeded. If, during operation, the automatic safety system does not function as required, the licensee shcIl take action as stipulated in the specification, which may in:1ude shutting dwn the reactor, notify the Cownission, review the matter, and record the results of the review including the cause of the condittoo aqd basis for corrective action taken to preclude i recurrence. I I
'L .
H-3 i
- l (iii) Limiting Conditions for Operation (LCO) Limiting conditions for operation -
l are the lowest functional capability or performance levels of equipment required for safe operation of the f acil'ity. When an LC0 is not met, the licensee shall shut down the reactor or follow specified remedial action as stipulated by the specifications to place the f acility in a safe condition until the operational limit or conditions can be met. The licensee shall notify the Cocynission, review the matter and record the results of the review including the cause of the condition and the basis for corrective action taken to preclude recurrence.
(e) (1) This section does not modify the technical specifications included in any license issued before (180 days af ter the effective date of this amendment). A license which does not contain technical specifications is deemed to include the entire safety analysis report as technical specifications.
i 1 (2) At the initiative of the licensee, any license may be amended to include technical specifications of the scope and content which would be required if a new license were being issued. ,
Rulemaking "Other Rule Changes" At least two dozen parts of 10CFR have references to Technical Specifications.
Each of the parts must be examined with the new understanding of what is or is not 4
proper for inclusion in Technical Specifications.
Appendix A Table A-1, ennumerates these CFR parts. The first and rrost important -
- change is 10CFR 50.36 and has been discussed above. Other references (10CFR 50.36a,10CFR50.43, etc.) specificy for inclusion in technical specifications, items which may not meet the recomended criteria and are not necessary to be 1 included in technical specifications.
~Several sections of the Code duplicate or overlap with specifications currently in the technical specifications (for example:'10CFR50, Appendices I,J,K,H, and R). In the new understanding of the regulatory basis for technical specifications, it becomes clear that only items meeting the criteria should be in technical specifications and that there is no need to place anything in both regulations and technical specifications. Federal regulations are requirements, there is no need
, to replicate such matters in individual technical specifications. Similarly, if items meet the Criteria and are included in technical specifications, they need not also be included in regulations, i
1 i
i l
I
1-1 Appendix I Probabilistic Methodology -
, Prcbabilistic methodologies for technical specification evaluatic, and improvement, which are under continuing development and are beginning to gain acceptance throughout the nuclear comunity, have also been considered by the Subcomittee, i While these methodologies appear very promising, the present licensing basis for
, nuclear plants is deterministic, being directly based on FSAR accident analyses which do not rely on probabilistic methods to demonstrate acceptability.
' To support NRC's use of probabilistic methods for technical specification improvements, the NRC's Office of Research is developing and providing infomation to the Office of Nuclear Reactor Regulation (NRR) regarding the use of
, probabilistic techniques to make decisions in the technical specification area.
- This program is known as the Procedure for Evaluating Technical Specifications i
(PETS). The NRC's prime contractor on this project is Brookhaven National L abor atory. Their objectives are to:
o Develop and demonstrate approaches incorporating risk and reliability insights for detemining A0Ts and STIs in the technical specifications; .
O Develop and demonstrate approaches for granting extensions to present technical specifications; and o Develop a procedures guide for determining acceptable A0Ts and STIs..
The industry has utilized the probabilistic evaluation approach in evaluating the l present technical specification requirements. For example, the Westinghouse Owners Group issued WCAP-10271, " Evaluation of Surveillance Frequencies and Out-of-Service Times for the Reactor Protection Instrumentation System" NRC review has been completed allowing the following technical specification revisions:
Change channel tests from monthly to quarterly; ~
c Change allowed out-of-service time from 1 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />; -
Change the time that a channel may be bypassed to allow ' testing of another j channel from 2 to 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />; and Allow routine channel testing in the bypass mode (with sune restrictions),
i The BWR Owners Group has an on-going program for development of a methodology' using PRA techniques in justifying STI and A0T extentions. NEDE-30851P, dated May 1985, -
. utilized this methodology in justifying SI extentions from monthly to quarterly and i A0T extentions from 1 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> for reactor protection system instrumentation. ,
i NRC review of NEDE-30351P is scheduled for completion by the end of 1985. The
{ report documenting the generic application and the demonstration cases for ECCS l Instrumentation, NEDE-30936P, is scheduled for submittal to the NRC in Octobe ,
1985.
I All four Owners Groups are planning to utilize probabilistic methodology on I specific technical specification improvements.
i i
1
I-2 EPRI also has a program on reliability and risk-based evaluation of technical sp cifications. The development of a theoretical foundation for a method of reliability-based technical specification changes has been completed (EPRI RP 2142). Additionally, the SOCRATES code has been developed and has been used on two utility applications. A third application is underway at this time, not to validate SOCRATES, but to critically evaluate the usefulness of the method. This should be completed in 1985.
- In early August 1985, the NRC and the AIF Technical Specification Improvement Subcormittee jointly sponsored a 2-day meeting for the exchange of information on the development and applications of probabilistic methods for technical specification improvement. The meeting participants included representatives frm the NRC TSIP, RRAS, and Office of Research, and NRC contractors on the PETS program; from the industry side the AIF, EPRI, Owners Groups, various utilities and other interested parties were present. The status of various NRC and industry activities was discussed.
Sane specific areas of imediate significance in the application of probabilistic ,
methods to technical specification improvement which were discussed during the meeting include:
Applications of generic analyses in support of plant-specific technical 1 specification change requests; Criteria for acceptability of proposed changes (including those which envision risk trade offs);
The use of cumulative risk as a measure for defining the limits of potential Technical Specifications change acceptability; The concept of cumulative allowed outage time for systems and components, and requirements for its implementation; and The future potential applications of probabilistic methodologies and their implications on licensing bases, data collection requirements, etc.
The industry and NRC representatives at the August meeting agreed that a continuing dialogue on the development of methods, procedures and criteria for application of probabilistic methods to technical specification improvement should be maintained.
The industry representatives requested an opportunity to review the products of the PETS program as they become available. In return, industry representatives, through the AIF Technical Specification Improvement Subcomittee, agreed to provide comments on these products and other issues to the responsible organizations within NRC.
M Regarding the need to ablish cumulative allowed outage times in the technical sptcificatioc( it i noted that no problem has been identified with the existing process that requir[eJthis regulatory action. Additionally, it is noted that probabilistic methodologies address anticipated future events, not historical pcrformance. The Working Group noted that considerable administrative detail would be required. It/isnotclearwhattimes(component,systemetc.)havetobe tracked. It isdot clea'r what action would be required if the allowed outage time is exceeded. But more importantly, additional emphasis on returning equipment to operation too quickly may compromise quality repairs in order to minimize expenditure of outage time allowances resulting in questionable benefit to the i overall safety of the f acility.
~' '
. Appendix J t.poendix J - Abbreviations and Acronyns
~
Act - Atomic Energy Act of 1954, as amended, 42 USC 2011, et seg ALARA - as icw as reasonably achievable AIF - Atomic Industrial Forum ,
ANPR -
advance notice of proposed rulemaking
- ANS ,American Nuclear Society A0T - allowed outage time -
ASEP - Accident Sequence Evaluation Program ATWS -
anticipated transient without scram BCL - Battelle Coltsnbus Laboratories BNL - Brookhaven National Laboratory
Combustion Engineering CFR - Code of Federal Regulations DBA - Design Basis Accident ECCS - emergency core cooling system EPRI - Electric Power Research Institute ESF - engineered safety features FR -
' Federal Regist:k FRANTIC -Formal Reliability Xnalysis including Normal Testing Inspection and Checking GE -
General Electric Company _ '
GENLTR - generic letter INP0 -
Institut'e of Nuclear Power Operations LCO - limiting . condition ~for operation NRC - Nuclear Regulatory Comission NRR - (Office of) Nuclear Re5ctor Regulation, NRC NTOL -
Near Tenn Operatino, Licensee PACRAT - Probability Analysis Code with Repair and Testing PETS -
Procedure for Evaluating Technical Specifications PRA - nenhabilitfic risk attettment
t 1 4 .- .
l RCIC -
reactor core isolation cooling j RES -
(Office of) Research, NRC RPS -
reactor protection system RV -
reactor vessel
- STI -
surveillance test interval STS -
standard technical specifications Tech Spec -
technical specification TSIP -
Technical Specification Improvement Project
- USC- -
United States Code ,
USQ -
unreviewed safety question (10CFR 50.59)
W -
Westinghouse Electric Corporation f
4
- - - _ - - - - . _ , _ . _ . - , . _ _ -..- _- . ,--