ML20003C757
| ML20003C757 | |
| Person / Time | |
|---|---|
| Issue date: | 01/29/1981 |
| From: | Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Ahearne J, Gilinsky V, Hendrie J NRC COMMISSION (OCM) |
| Shared Package | |
| ML20003C742 | List: |
| References | |
| FRN-45FR65474, REF-10CFR9.7, RULE-PR-50, RULE-PRM-50-29, TASK-OS, TASK-RS-220-5, TASK-SG-029-3, TASK-SG-29-3 NUDOCS 8103180209 | |
| Download: ML20003C757 (41) | |
Text
'
'npy b]l.44 A
\\(
3 UNITED sTA TES t
[ g, c(y j NUCLEAR REGULATORY COMMIS$10N wAsmwoTon. o. c. cosss
- g jg
\\"..CM/
JAN 2 91931 MEMORANDUM FOR:
Cheirman Ahearne Comis.tioner Gilinsky Cor.issioner Hendrie Comissionar Bradford FROM:
William J. Dircks Executive Director for Operations
SUBJECT:
PROPOSED RULEMAKING TO AMEND 10 CFR PART 50 CONCERNING ANTICIPATED TRANSIENTS WITHOUT SCRAv. ATWS) E'isNTS (SECY 80-409) i As requested in the December 9, 1980 memorandum from S. J. Chilk (M501125),
the Federal Register Notice has been rewritten. Enclosure "A" to SECY S0-409 should be replaced with the revised Enclosure "A."
The discussion in the Notice has been rewritten to:
1.
more clearly state in the History section the shift in aporoach between prevention ano mitigation, in response to Conmissioner Gilinsky's request; 2
revise the conclusion in the Basis for the Procosed Rule tu clarify the discussion of the acequacy of tne reilanliity of the reactor protection system, in response to Comissioner Hendrie's question; and,
~
3.
clarify in the Content of the Procosed Rule section that the requirements of Ine proposeo section 50.49 will serve as the exclusive regulatory basis for the imposition cf require-ments for ATWS events in respense to coments by ine General Counsel at the November 25 meeting.
turginal-lines in the revised Enclosure "A" indicate the locations of the changes made from the November 7 version.
In response to Co:rtaissioner J$ndrie's question regarding use of reactor head vents for' safety valving, a discussion of possible means of providing additional relief capacity is provided in Enclosure "M."
The Federal Register Notice also has been revised to show in comparative text form how the alternatives presented in Chairman lhearnels cemorandum to W. Dircks on December 10, 1980 could be incorporated into the proposed M.
~
Contact:
W. Minners, NRR 49-27581 180 hb
~
The Commissioners 2
1.
Old Plants. The Chairman's memorandum suggests requiring only Alternate 2A, but does not indicate any changes to be made in the proposed rule. The staff interprets this to mean that the 'st:ff's suggestion (i.e., do not explicitly address these p. ants in tne rule but consider exemptions from the rule based on analysis by the licensees and evaluations similar to the systematic evaluation program) is to be adopted. This interpretation requires no changes to the proposed rule.
If old plants are to be specifically exempted, the simple statement:
" issued an operating license efter August 1969" could be inserted in paragraph b (1) of the proposed rule, and the explanatory paragraph in the Content of the Proposed Rule deleted.
2.
All Westingh:ase Plants. The Chairman's memorandum suggests requiring Asternate 2A plus Mitigation Systems Circuitry for all Westinghouse plants and elimination of the additional requirements for post-84 plants. The staff believes that the proposed rule, as written, would only require that diverse mitigation systems circuitry be installed on Westinghouse designed plants. However,'the staff recommends that Westinghouse plant owners should demonstrate by analysis that their plants meet the acceptance criteria. The present generic analyses should be adequate, except in the area of long-term cooling.
Enclosure A reflects this position and has not been revised.
The Chairman's memorandum and enclosure are inconsistent as to whether Westinghouse plants should or should not have a supplementary protection system (SPS). The memorandum suggests Alternate 2A, which includes an SPS.
However, the revitt:d rule attached to the memorandum left the exemption from the requirement for an SPS in the rule.(Paragraph C (2)). The staff has adopted the ACRS recommer,dation for Westinchouse designed plants (i.e., for pre-84 no requirement to have an SPS).
Enclosure A reflects this position and has only been revised to delete the requirement that the analysis of Westinghouse designed plants be done with a more conservative moderator coefficient.
As suggested, Enclosure A has been revised to delete any additional requirements for post-84 plants. This includes no requirement for an SPS in post-84 Westinghouse designed plants, which is contrary to the ACRS recommendation for these plants (i.e., require an SPS).
3.
CE and B&W Flants. The Chairnan's memorandum suggests requiring Alternate 2B, or plant specific analysis to demonstrate acceptable consequences for all CE and B&W plants. Therefore, the only revision made to Enclosure "A" is that the additional requirements for post-84 plants have been deleted.
4.
BWR Plants. The Chairman's memorandum suggests requiring only Alternate 28 for all BWR's.
Therefore, the' only revision made to Enclosure "A" is that the additional requirements for post-84 have been deleted.
The Commissioners 3
5.
Acceptance Criteria.
The Chairman's memorandum suggests changing the ASME Level C stress limits to Level D plus strain limits on components, and a demonstration of their operability.
Enclosure "A" was revised to show this change.
However, the staff believes t' hat the Level C Service Limit is a better criterion, as discussed in Enclosure L, and that any reduction in conservatism is better obtained through revisions in the evaluation model, primarily the moderator temperature coefficient.
The staff is preparea to discuss the desirability of the above changes at the next meeting with the Commission.
High Population Sites:
The Chairman's memorandum also requested that the staff address how special consideration could be given to plants near popu-lation centers.
The staff's intent in including the language at the top of page 17 of the Notice regarding plants determired in the future to be at high risk sites was to implement the post-84 plant requirements, i.e., a more conservative evaluation model and the inclusion of single failures.
The ACRS in their April 16, 1980 letter on NUREG. <460 made recommendations for specific piants (see En:losure "G", SECY-80-409).
The Chairman asked that the views exp essed in R. Minogue's memorendum to the Chairman on A WS calculations be reconciled with this proposed rule.
R. Minogue's memorandum to the EDO at Enclosure "N" recommends that the rule be published as it stands and that a limited rule option for BWR's in MARK I and 11 containments, along with other options, beconsidered after public comment is received.
William J. Dircks Executive Director for Operations
Enclosures:
A - Revised Notice of
- Props'.ed Rulemaking L - Proposed ATWS Rule, Acceptance Criterion, Primary System Pressure M - Conceptual Designs for Additional Relief Capacity -
N - Memo Minogue to Dircks, dtd-1/28/81, " Proposed ATVS Rule cc:
OPE 03C SECY e
M4
ENCLOSURE A l
I 6
e e
c (Draft 12/18/80, NUCLEAR REGULATORY COMMISSION 10 CFR PART 50 DCMESTIC LICENSING OF PRODUCTION AND UTILIZATION FACILITIES Acceptance Criteria for Protection Against Anticipated Transients Without Scram (ATWS) Events for Light-Water-Cooled L'uclear Power Plants AGENCY:
U. S. Nuclear Regulatory Ccmmirsion, ACTION:
Proposed Rule.
~
SUMMARY
Following extensise study of anticipated transients without scram (ATWS) events, the NRC has concluded that the probability of an ATWS event occurring, as well as the attendant consequences of such an event should one occur, are unacceptably high.
Accordingly, the NRC is considering amending its regulations to require in.provemen's in the design, construction, and t
operation of nuclear power plants in order to (1) reduce the probability that a nuclear power _ plant would fail to scram (rapidly shut down) following the occurrence of a transient event (i.e., an abnormal operating condition), and (2) mitigate the consequences of a failure to scram following a transient, should such an event occur.
DATES:
The comment. period expires (90 days after publiestion),
ADDRESSES:
Comments should be submitted.in writing to the Secretary of. the Commission, U. S. Nuclear Regulatory Commission, Washington, D'. C. '20555,
7 -.. - _ _ _ -.
i i
3
[
Attention:
Docketing and ServL:e Branch.
All comments received will oe
(
available for public inspection in the Commission's Public Document Room at i.-
l 1717 H Street, N.W.,' Washington, D. C.
I t
O O
O e
O 4
0 t
G e
W 6
e O
+
e e
e
'FOR FURTHER INFORMATION CONTACT:
'i Medhat M. El-Zeftawy, Office of Standards Development, U.S. Nuclear Regulatory Commission, Washington, D.C.
20555, (301) 443-5921 SUPPLEMENTARY INFORMATION:
History: The question of whether and to what extent anticipated transients without scram (ATWS) events should be considered in the design and safety evaluation of neulear power plants has been the subject of extensive and continuing studies by' the NRC staff and the regulated industry.
It has long been recognized that, should an ATWS event occur, the pote-tial for release of a substantial amount of ' radioactive fissien products as a result of the melting of reactor fuel is significant.
It was not until 1969, however, that it became apparent that the probability of such an event occurring may in fact be higher than earlier analyses indicated.
Early that year, a consultant to the Advisory Committee on Reactor Safeguards (ACRS) pointed out that the reliability of the reactor protection system ray be diminished as a result of a common mode failure (i.e., the failure of cultiple components due to a commen single cause).
TThe concern was that an extraordinarily high reliability of the reactor trip and reactivity _ shutdown systems was required, considering the relatively high i
rate of challenge by anticipated trant ents, the increasing number of nuclear i
power plants, and the desire'to assure that the potentially severe consequences l
of failure were very unlikely.
Attaining such a high reliability requires that the frequency of occurrence of common mode failures be extremely low.
This extremely low frequen.cy was lower than coul'd be confidently predicted a
D
~
2 by currently available reliability assessment methods.
Thus, the practicality of attaining and demonstrating the required high reliability in a single reactor protection system subject to common mode failures was questionable.
Because of this question, analyses of the consequences of postulated ATWS events were requested from reactor designers.
After reviewing these preliminary and simplified analyses, the staff confirmed that the consequences of ATWS events cou;i be severe in that several anticipated transients would require prompt action to shut down the reactor in order to. oid high pressure in the primary system ano possible offsite effects.
The staff's preliminary results on ATWS were discussed with the ACRS in September 1970.
In August 1971, the ACRS and the regulatory staff concluded that a design change to the proposed Newbold Island boiling water reactor units was appropriate to limit the possible consequences of ATWS.
In April 1972, the staff transmitted to the ACRS a proposed set of positions and actions to be taken to implement the conclusions of the staff and ACRS studies on ATWS.
In January 1973, as a result of further review and discussion, the staff transmitted to the ACRS an amended proposed position on the need for protection against ATWS.
The ACRS responded in April 1973, agreei'ng with the amended position.
In September 1973, the staff published the " Technical Report on Anticipated Trans'ents Without l
Scram for Water-Cooled Power Reactors" (WASH-1270)* containing its position on ATWS.
t t
x Copies of these reports may be purchased from the Division of Technical
.Information and Document Control, U.S. Nuclear Regulatory Commission Washington, D.C.
20555.
i-
3 In this report the staff proposed different requirements for different plants.
Plants for which construction permit applications were co.keted from early 196S to October 1976 were to be required to incorporate the design changes necessary to assure that the consequences of a postulated ATWS event would be acceptable, that is, these plants would have the capability to mitigate ATWS events.
These plants were also to be required to modify those portions of the reactor protec- -
tion systems that might be particularly vulnerable to common mode failures in order to prevent some ATWS events.
Plants for which CP applications were docketed on or after October 1, 1976 were to be required to incorporate design changes that would significantly improve the reiaaaility of the reactor protectica system such that ATVS events were more likely to be prevented.
This was to be accomplished by adding a reactivity shutdown system which was to be diverse and independent from the reactor trip system (the sensors, logic and scram breakers) and the control rods drives, but not from the control rods themselves.
I t.
In November 1975, (SECY-75-668) the staff proposed and the Commission agreed to treat both classes of plants identically and require that both classes of plants have the capability to mitigate ATWS events and modify the portions of i
the reactor c otection syste= that were particularly vulnerable to common mode failures.
Although a diverse reactivity shutdewn system was no longer to be required, it was not prohibited.
Since the pra:tical implementation of a diverse reactivity shutdown system would either be a. boron injection system or r
L __
4 the replacement of some rod drives by diverse drives, evaluation of the performance of the diverse reactivity shutdown system against acceptance-criteria would be required as with any mitigating system.
After WASH-1270 was issued, reactor manufacturers in conjunction with the staff, began to develop acceptable methods of performing analyses of AWS events.
A draft industry standard, ANSI-N661, was written, which outlined genera.1 guide-lines for the analysis of AWS events in PWRs.
In October 1974, each manufacturer of nuclear steam supply systems (NSSS vendors) submitted reports describing the analysis of AWS events for their reactor designs.
The staff reviewed thitse reports and in December 1975 issued status reports with evaluations of these analyses.
The NSSS vendor transient analysis methods were generally acceptable except for the treatment of system failures and some system parameters.
St.b s e-quently, in mid-1976, applicants were requested to perform analyses for their plants using the methods developed'by the NSSS vendors, modified as indicated i
in the staff status reports.
In 1975, contemporary with the status reports, the staff published the Reactor Safety Study (RSS), WASH-140'J*, which contained estimates of the probability and consequences of core-melt from various causas including AWS events.
9 In 1976, the Electric Power Research Institute (EPRI) published a set of' reports which had been submitted earlier and provided their assessment of the signifi-cance of ATWS events and supporting analysis and data.
As a result of criticism of the staff position and the new information submitted by EPRI and publisned L
5 in the RSS, the staff reviewed and evaluated the information then available on the subject with particular emphasis on the material developed subsequent to the publication of the status reports.
Th'e results of their review were published by the staff in April 1978 in the report " Anticipated Transients Without Scram for Light Water Reactors" NUREG-0460, Volumes 1 and 2*.
In this report, the staff recommended that the previous position be modified.
The staff still concluded that nuclear power plants should incorporate mitigating systems to assure that the consequences ci STWS events would be acceptable, but the requirement to make changes to the reactor protection systems was dropped.
The basis for the conclusion remained the same:
the reliability of reactor protection systems, and therefore the frequency of ATWS events resulting in cere-melt, CNld not be shown to be, a small contribution to the overall risk from nuclear power plants as estimated in the Reactor Safety Study.
This report was discussed with the ACRS in a series of meetings through the Fall of 1978.
Late in 1978 the Risk Assessment l
Review Group reported to the NRC on its assessment of the RSS and the current i
state of risk assessment methodology.
This group recommended that, in general,
+he use of probabilistic risk analysis methodology be avoided for the determina-tion of absolute risk probabilities for subsystems unless an adequate data base existed and it were possible to quantify the uncertainties.
The Nuclear Regula-tory Commission accepted this recommendation an'd included it in a statement of policy issued in January 1979.
Based on the recommendations of the Risk Assessment Review Group; a review by the staff Regulatory Requirements Review Committee; and the information provided l
~
6 by nuclear' utilities, architect-engineering, and reacter manufacturers at the 1978 ACRS meetings; the staff reexa::ined the approach recommended in.NUREG-0460.
The results of this reexamination were published in Volume 3 of NUREG-0460 in December 1978.
In this report the staff concluded that the safety objective presented in the report was not satisfactory for use in regulatory decision making, but that engineering evaluation and judgement, supported by quantitative risk evaluation, should be used to determine the appropriateness of the various alternative plant modifications described.
The staff recommended specific modi-fications for different classes of plants.
The staff recommended that early operating plants provide a diverse reactor trip system that would help prevent ATWS events resulting from common mode failures in the electrical portion of the reactor protection system.
The staff further recommended that the other
~
operating plants and plants under construction provide a diverse reactor trip system and the additional systems necessary to mitigate the consequences of most ATWS events.
For new plants, the staff recommended that systems be provided to mitigate the consequences of almost all ATWS events.
j.
In February 1979 the staff requested that generic evaluations be performed by l
l the industry to confirm that these mcdifications would achieve the desired
[
objectives.
l The accident at Three Mile Island forced deferral of all NRC and most industry
- work on AWS and the information submitted fell' far short of the reques The l
information that was received did not confirm that the proposed modifications i
would achieve the objectives.
In Volume 4 of NUREG-0460, published in March 1980, i
l the staff revised its recommendations and proposed to require some specific modifications to plants immediately and to impose further requirements after W
7 conducting' a rulemaking proceeding and issuing a regulation.
The specific modifications included the addition of a diverse reactor trip system and'the upgrading of some installed systems, primarily by making the actuation circuitry diverse from the rcactor protection system, to increase the capability to miti-gate ATWS events.
The proposed regulation would require all but the early operai.ing plants to have the capability to mitigate the consequences of almost all ATWS events in addition to the specific modifications.
Subsequently the staff decided to recommend that ali ATVS requirements be imposed by regulation after a rulemaking proceedit.g.
Basis for Proposed Rule:
The statutory basis for deciding whether, and to what extent, ATWS events should be considered in the design and safety evaluation of nuclear power plants is set forth in Section 1611(3) of the Atomic Energy Act.
That section grants to the Commission the authority to " prescribe such regulations or orders as it may deem necessary... in order to protect health and to minimize danger to life or property."
Implementation of this direction ir. the regulations for nuclear power plants has been based on dual objectives, prevention of accidents and mitigation of their consequences should they occur.
Thus, conservative design, construction, cnd operation of plants are required so that the accidents will be prevented (i.e., have a low probability of occurrence).
Then, to provide defense in depth, the capability to mitigate the consequentas of accidents that are postulated to occur is required even though the design includes measures to prevent them.
In specifying the requirements for preventing or mitigating accidents, not all
~
accidents that can possibly~ occur are postulated to occur.
One function of
8 the regulations and requirements of the NRC is to specify which possible acci-dents and consequences are sufficiently probatle to warrant preventive or citigative measures.
It is within this framework that the NRC has concluded that the probability of ATWS events ocurring over the lifetime of light-water-1 nuclear power plants and the potential magnitude of censequences arising from such events, should they occur, are sufficiently great to warrant the impositien of requ:rements designed to reduce the probability and mitigate the consequences of ATWS events.
The review and evaluation by the NRC staff of the informatio-that has been developed over the past ten years on ATWS events and of the manner in which they should be considered in the design and safety evaluation of nuclear power plants that form the basis for this conclusion is contained in the report
" Anticipated Transients Without Scram for Light Water Reactors," NUREG-0460, Volumes 1 through 4.
There are two primary factors in the staff's evaluation.
The first is the degree of assurance that ATWS events can be prevented, which depends on the reliability of current reactor protection systems.
The second is the capability of existing reactor designs to mitigate the consequences of ATWS events.
The reliability of current reactor protection systems has been estimated based on the operating experience to date and'reliabiiity analyses.
However, the very high level of reliability required is difficult to demonstrate with confidence because it depends en accurately determining the rate of common mode failures.
Common mode failures involve failures of multiple components resulting frora a single cause or event.
Reactor protection systems are carefully reviewed
9 to identify and eliminate all but the most unlikely common mode failures.
However, one common mode failure in the reactor trip portion of the protection system of a comearcial nuclear power reactor has occurred during approximately 700 reactor years of operating experience.
The failure was detected during normal surveillance and corrected before any event requiring a reactor scram occurred.
~
Common mode failures have also occurred in other systems in nuclear power plants and other potential common mode failures in reactor protection systems have been identified.
Because of the low rate of occurrence of common mode failures, operating experience is not, and cannot be, sufficient to cone.lusively determine on a statistical basis whether reactor protection systems are reliable enough to make the probability of unacceptable consequences from ATPS events acceptab1y small.
Reliability analysis methods are also inadequate because they must treat common mode failures either in an arbitrary manner or use tae highly uncertain estimates of common mode failure rates derived from operating experience.
While qua'ntitative estimates of protection system reliability provide important infor-mation, the conclusion as to the adequacy of protection system reliability must be based on enginecting judgment.
The NRC has concluded tnat the reliability of current reactor protection systems has not been demonstrated to be adequate and most likely is not adequate.
The probability of severe consequences resulting from ATWS events is also affected by the capability of nuclear power plants to mitigate ATWS events.
This capability varies depending on the design of the reactor system, and the
~
status of systems and the value of system process variables at the time the e
10 t
event occurs.
The capability of a plant to citigate ATW5 events can be assessed by analyses.
However, uncertainties in the design characteristics of the reactor, the probability of failure of the citigating systems and the probability that the values of syste: process variables will be different frc: these assu ed in the analyses, all combine to produce uncertainty in the results.
Therefore, the difficulty in demonstrating a capability to adequately citigate ATWS events is sitilar to the difficulty in demonstrating that ATWS events can be prevented.
However, based en analyses performed to d' ate, it is clear that in most cases present reactor designs have inadequate capability to citigate the consequen:es of many postulated ATWS events should they occur.
Content of the Preposed Rule:
Having concluded that improvecents to reduce
~
the probability of severe consequences frc: ATW5 events should be cade, the staff developed four alternative sets of require ents that wculd provide increasing reduction in this probability and would require increasing amounts of codifications.
The alternatives were first described in Volume 3 of NUREG-0450 and again in slightly revised form in Volume 4.
The intent of the proposed rule is to adept a combinatien of the alternatives rece ended in Volume 4 (except for one change for reactors designed by Westinghouse and licensed to operate before 1984).
The preposed rule would also implement the requirements in a different manner from that described in Volume 4 of NUREG-0450.
The for: of the requirements in the prc' posed ru'le are also different frc= these reco:: ended in NUREG-0450 in that the preposed rule specifies acceptance criteria
- for ATWS mitigating syste=s while the required citigating systems are specified in Volume 4.
f 11 Alternative 1 is to make no modifications at all.
As discussed, the NRC has concluded that the reliability of current reactor protection systems is 'insuf-ficient and.that the probability of ATWS events is sufficiently great to warrant improvements.
Therefore this alternative is not represented in the proposed rule.
Alternative 2, as modified in the proposed rule, would increase the eeliability of the reer. tor trip portion of reactor protection systems and improve the capa-bility of existing systems to mitigate some ATVS events.
Reliability of the reactor trip systems would be increased by the addition of supplementary protec-tion systems that would be independent and diverse from the reactor trip portion of the current reactor protection systems.
Diversity would be achieved by the use of components from different manufacturers; by the use of components having different principles of operation, or power sources; and by the use of components in different operating modes (normally energized vs. normally deenergized).
This alternative would not provide increased reliability of the reactivity control portion of the protection system, i.e., the control rods and control rod drives.
However, in the case of reactors designed by General Electric it was proposed to increase the reliability of a portion of the control rod drive system, i.e.,
the control rod drive scram discharge volume. The capability to mitigate ATWS events would be improved by providing actuation circuitry that was diverse from the reactor protection system for some existing systems such as primary system relief valves, turbine trip, and auxiliary feedwater in PWRs and the. recircu-lation pump trip in BWRs.
12 The staff proposed in Volume 4 to implement only Alternative 2 for the ten older plants which began operation before late 1969.
Because of their unique'charac-teristics, the staff believed that more extensive modifications would net be appropriate for these plants.
The proposed rule does not explicitly address these plants (except in the implementation schedule), but the intent is to consider any exemptions from the acceptance criteria ~ of the proposed rule for these older plants based on analyses by the licensees and evaluations similar to those conducted under the Commission's systematic evaluatio.n program (SECY-77-561 October 1977) in context with the overall safety of these facilities.
Alternative 3, as modified in the proposed rule, would increase the reliability of the reactor trip portion of the reactor protection system for some plants and provide for the mitigation of most ATWS events.
The reliability of the protection system would be increased in the same manner as in Alternative 2.
However this increased reliability of the reactor protection system would not be required in plants that have a greater capability to mitigate ATWS events.
The mitigation of most ATWS events in PWRs was expected to be accomplished as in Alternative 2, except that means would be required to isolate the containment early in an ATWS event upon detection of radiation released from failed fuel.
The mitigation capability of BWRs was expected to be increased by providing automatic initiation and increasing the flow capacity of the Standby Liquid Control System.
Considering the state of design and construction, and a balanc-ing of pubic safety benefits against economic cost, the staff concludes that all plants nee 4v4eg-en-epentieg,14ceese-Mwe-je64 should be required to implement Alternative 3 as modified in the proposed rule.
=
wmW
1 13 I
s l
Alternat?we A -as. cod?tted.?n-the-psspassd-suls, would i= crease tbe-raliability r....
8!".I '.".5'.aejeP-tPjp7pertjee,e{7jhe,geeejer-pgetectien-system-e{-e44-p4ents-and p rev ?de-f e r-t he-mi* ' gatie e.-e f.vi r tee 44y.eii-ATW5-events:--The-re44 eb444 ty.ef--
t h e -p ys te s t ? s s-sy s tem s -we si d.b e-4 ner e as e e-48-the-s ame-maa ne r-es-4 e-Al ternet49 e-Er The.n?tiga tien.e f.virt wa44y.a44.ATWE-events.w.as-expected-te-require-s 4 ge4(4 cent design-shanges,--The.mit4gatien.eapability.ef.pWRs.w.as-expected.te-be-sehstan-tially.facreased-by.additismal-presswre.ve44ef-espas4ty-4n-the-reacter-see4 ant system.--!ha-m4tiga tio n-cap abi;4 ty.e f.BWRs.w.as. expected-ts-be-4 eereased.by-the additic -st-hightcapacity.seutven-psissa-inf.estisa-systens,..In.ba4anciR5-pub;4s-satsty.bs:sfits.against-esc smic-scst -the-staff-conc 2wded.that-these-extens4ve -
3.................
desig:. changes.could-caly bs-practica22y.insc parated fa plaats-act-near-comp 2e=
tion acd mot to be lics: sed bsfcre 1984 The proposed requirements in Volume 4 of NUREG-0450 were in the form of specific design changes.
The proposed rule also explicitly specifies the design changes required to improve the reliability of the protection systen and the response for containment isolation, but the changes in mitigation capability are required through the specification of acceptance criteria, criteria for evaluation models, and mitighting system design criteria.
The specification of criteria requires licensees and applicants to demonstrate that the design of their plant is in compliance and thus provides more atturance that the safety objective is being attained.
This form also allows the designer more flexibility in design and a greater potential for optimizing costs.
Although the ultimate safety. objective is to limit the release of radioactivity to the environment, the acceptance criteria in the proposed rule are directed e
l 14 toward assuring the integrity of the reactor coolant system and the reactor core following ATWS events.
The staff recognizes that failure to, satisfy these acceptance criteria does not necessarily result in severe radiological conse-
~
quences and has considered the additional safety margin in developing tne pro-posed rule.
In formulating the proposed rule, the Comission has considered the potential applicability of Part 100 offsite radiaticn exposure guidelines.
to the generic ATWS issue.
Upon careful review of the scope and intended purpose of Part 100, the Commission has determined that the guidelines con-taineo therein provide an inadequate framework, both from a legal and technical perspective, within which to impose requirements for the prevention and mitigation
~
of ATWS events. Accordingly, the Commission has decided that proposed section 50.49 will serve as the exclusive regulatory h sis for the imposition of require-ments designed to prevent the occurrence of ATWS events and, should they occur, j
mitigate the resulting consequences.
If only the guidelines of Part 100 for calculated offsite du3es were specified, the flexibility for the designer would be increased, but the attainment of the safety objective would be more difficult ot demonstrate.
If systems designs were specified, the flexi-bility of the designer would be reduced, and the demonstration that the safety objective were attained would be generic rather than for specified plants.
Prior attempts at such a generic demonstration have-been unsuccessful, as discussed above.
The level of-safety, whether the mitigation of most or virtually all ATWS events, is specified th' rough the criteria'for acceptable evaluation models. Since the parameters in the evaluati a model are all uncertain to some degree and some vary over the-lifetime of the plant, the level of safety is determined to a large extent by the degree of-conservatism in the parameters used in'the evalua-
~
tion models, which affect the conservatism of the calculated consequences of A
l 15 postulated ATWS events. The proposed rule specifies that realistic values of parameters may be used when the value is known with reasonable accuracy, but that parameters with large uncertainties must be conservatively treated.' The intent is to obtain realistic analyses of the course of ATVS events, yet predict the consequences conservatively.
In order to assure that the consequences of cost ATWS events will be within the acceptance criteria, the proposed rule specifies that the value used for parameters that vary over the lifetime of the plant (the most significant of these is the moderator temperature coefficient) roast be a vales that is not exceeded over most er-y-fMee4b-e44 of the plant lifetime.
In the case of the moderator temperature coefficient, the value used in the evaluation model that was less negative than tha value expected to be experienced during 90 e~c9 percent of the design lifetime of the plant would assure that the consequences of most ss-v4stwa24y-a44 ATWS events would not violate the acceptance criteria.
Although improvements in the capability to mitigate ATWS events provide a signif-icant increase in the level of safety, there is some uncertainty associated with this conclusion.
This uncertainty derives from the uncertainty in the reliability of mitigating systems and in the evaluation models used to define them.
Because of this uncertainty the staff believer that improvements in reactor protection system reliability should also be required.
Such modifi-cations to present reactor protection systems, as with any modifications to a nuclear plant, have the potential for introducing unrecognized failure modes that could result in a decrease in the level of safety.
A careful design process in conjunction with the quality assurance, verification, and test programs is necessary to assure that this will not occur.
However, the implementation of h
these improvements in reliability in some plants is to be accomplished within
~
16 two years, and such a short design and installation schedule might compromise the design program.
In plants, such as those designed by Westinghouse, which have a capability to mitigate nearly all ATWS events and where the level of safety is already high, the Advisory Committee on Reactor Safeguards (ACRS) recommended omitting tre requirement for improvements in the protection system reliability. Thes ;.the prepese+reie..t.i.ie.w.s..th.e..pr.et.t.e.titm syste= imp cuer.ects.6 be-cr.Jtte6...4!-r+re-conse+ vet 44+.ve4*s.e f. the.W... ers.;
sucb.as moderatc: tempe:ature.cse!!icie&...are.used.4e-#e.evsAeties-one4eas and.the.capabi.lity.to-ce p4y-with.th-estepte :e--eriteris..is-G...wstreted---in-plents..i.i.ee.n.t.e.d..i.n..e...z.f.t.e.r..i.?.S.4. a.i.t.e..t.i.c.t..tvai.i.t.t.i.t..t.o..c.a.s.i. uT Inrin.3tt1.1..1.M.
?'d4NEutieas]{e]jpe gjggj {,gjeNs,y{jp,W,jp,tsyj$the-Mjp eresess.w.eeld-eet-be-cempremised.end-imp eve.me.nts-ire-the pd.eetien-systems ef a44.6(-these.plaats-45-required-by-tht prepesed.reie.
One plant modification that would be required by the proposed rule is already being implemented on boiling water reactors.
In an order dated February 21, 1980, licensees of EWR plants were directed not to operate after December 31, 1980 without a recirculation pump trip installed.
Licensees have also been directed (IE Bulletin No. 80-17 dated July 3,1980 and NUREG-0737, Clarification l
of TMI Action plan Requirements *) to assure that operating procedures and operator training address the actions to be taken in the :' ants as now designed if an ATWS did occur.
These equirements are prudent'mehsures which will reduce the risk from ATW5 events durin.7 the interim period before the plant modifications determined by the Commission to be necessary and included in a final rule, can be installed.
17 In particular cases, additional requirements or earlier implementation may be appropriate.
For example, candidates would be those existing nuclear po'wer plants that_are considered to be at high risk sites owing to a combination of population density, meteorological conditions and other factors.
Identification of these sites is a subject of another Commission action and any additional ATVS requirements for these units would be subsequently considered.
Concurrent with this publication of the proposed rule for comment, the staff is also publishing a proposed regulatory guide for comment.
This regulatory guide provides guidance on the evaluation models, mitigating system design requirements, and other licensing requirements.
Implementation of Reauirement_s:
The proposed rule provides for implementation of the requirements in stages in order to gain the greatest increase in safety in the shortest time ant at the least cost.
The modifications to improve the reliability of the protection system and the mitigating system actuation cir-cuitry would be required by July _1,1952.
In order to accomplish this, descrip-tions of the modifications are to be submitted for review by the NRC by July 1, i
1981.
Since these modifications involve instrumentation, control and logic j
circuits that are for the most part outside of the containment, most of the installation could be accomplished with the plant operating or during refueling outages and a two year design and instailation schedule appears appropriate.
i The proposed rule provides that implementation of any modifications required to meet ATWS acceptance criteria be completed'by January 1,1984 for pressurized
18
~
water reactor plants and for the oldest boiling water reactor plants.
All other boiling water reactor plants would be required to complete the modifications by July 1, 1982.
In order to accomplish this, evaluation models are to be sub-mitted for review by March 1981 and descriptions of the modifications are to be~ submitted for review by July 1981 (by December 1981 for the ten oldest plants).
Since the modifications to the boiling water reactors would consist of piping changes to the Standby Liquid Control System, which is primarily outside contain-ment, a two year design and installation schedule appears to be appropriate.
Since any modifications determined to be required for pressurized water reactors would likely require the installation or modification of valves on the pressurizer, tne more extended schedule for these plants is appropriate.
~
The NRC believes that the likelihood of severe consequences arising from an ATWS event during the period of thie implementation is acceptably small.
This judgment is based on a) the favorable experience with the operating reactors, b).the limited number of operating nuclear power reactors, c) the inherent capa-bility of some of the operating PWRs to partially or fully mitigate the conse-quences of ATWS events, d) partial ATWS mitigative capability of the recirculation pump trips feature which has been implemented on most operating BWRs and which is required to be implemented on the remaining BWRs by December 31, 1980, and e) the interim steps taken to develop procedures ar.d train operators to further reduce the risk from some ATWS events. 'On the basis of these considerations, the NRC believes that the implementation schedule in the proposed rule is acceptable and will minimize the risk of hasty modifications which may be counterproductive to safety.
2
~
19
~
Pursuant to the Atomic Energy Act of 1954, as amended, the Energy Reorganization Act of 1974, as amended, and section 553 of title 5 of the United States Code, notice is hereby given that adoption of the following amendments to 10 CFR Part 50 is contemplated.
A new Section 50.49 is added to read as follows:
550.49 Acceptance Criteria for Prctection Against Anticipated Transient Without Scram Events for Light-Water-Cooled Nuclear Power Plants (a) As used in this section:
(1) " Anticipated Transient Withcut Scram" (ATWS) means an antici-pated operational occurrence as defined in Appendix A of this part followed by the failure of the reactor protection system specified in General Design Criterion 20 of Appendix A of this part.
(2)
"ATWS evaluation model" means t'
- alculational framework for evaluating the behavior of tne nuclear power plant during a j
postulated ATWS event.
(3)
"ATWS mitigating systems'" means those systems including associated controls, instruments, power supplies and other systems assumed l
l to function when evaluating the behavior of the nuclear power l
plant following an ATWS event.
I
20
~
(b)
(1)
Each light-water-cooled nuclear power plant shall be designed, s
constructed, and operated such that the consequences of postulated anticipated transient without scram (ATWS) events calculated in accordance with an ATWS evaluation model, approved pursuant to paragraph (b) (2) of this section, conform to the following criteria:
(i)
Primary system pressu're.
The calculated reactor coolant system (RCS) pressure and temperature resulting from postu-lated ATWS events shall be limited so that either-(A3 the calculated maximum primary stress anywhere in the RCS pers-sure boundary does not exceed that permitted by the " Level E-D Service Limit" as defined in Article NS-3000 of Section III of the ASME Boiler and Pressure Vessel Code and the calculat,id deformat"on of RCS components is limited so that the 6parability of components necessary to safely bring and maintain the reactor at a cold shutdown condition is not impaired, er-(B3-the-integrity-er-eperabiiity-of-RES cempenents shaii-be-demonstrated-based-on-censervative as s e s sments-of-te s ts - c e nducte d-to-de t ermi ne-th e-i nte g ri ty or eperabiiity ef-cempenents-under-the cenditiens accompanying pesteinted-ATWS events and-based-on-the-iikeiy cenditien of-the-ceeper.ents ever-their-design-iifer
,/
Dbd y
J
21 9
(ii) Fuel integrity.
The calculated damage to the reactor core as a consequence of postulated ATWS events, including oscil-lations of power and flow, shall be limited to assure that the core geometry is not significantly distorted such as to impair core cooling or safe shutdown.
(iii)
Radiation Release.
The calculated release of radioactivity from the fuel rods to the reactor coolant system during postulated ATWS events shall not exceed one percent of the radioactivity within the fuel rods of a pressurized water reactor or ten percent of the radioactivity within the fuel rcds of a boiling water reactor.
(iv)
Containment.
Tne calculated containment pressure, tempera-ture, and humidity resulting from postulated ATWS events shall not exceed the design value3 of the containment struc-ture and components or the contained mitigating systems, equipment and components.
For boiling water reactor pres-sure suppression containments, the relief or safety valve discharge line flow rates and suppressica pool water temperatures shall be limited so tnat steam quenching instability will not result in destructive vibrations.
(v)
Long-term shutdown and cooling..The reactor design shall permit the reactor to be safely brought to and maintained W.
22 at a cold shutdown condition following postulated ATWS events without insertion of control rods.
(2)(i) ATWS evaluation models shall, with reasonable accuracy or acknow-ledged conservatism, represent the actual characteristics of the facility modeled and each significant physical phenomenon that would occur in the reactor and related systems during the course of the modeled event.
Evaluation models shall represent the effect of the failures in mitigating systems that are a direct consequence of the ATWS event being modeled.
[Fer facilities-issced-eperating-iicenses en er-after-dancary-i--1984 and net standardiced-te-a-facility at-the-same site-that was issced an-eperating-iieense-befere-daneary-1--1984-evalcatien medeis shaii aise represent-the effect-ef-the-iikely-randem singie-faiieres ef active cemponents-in mitigating-systems-]
j (ii) The value of parameters that vary over the lifetime of the i
facility or represent the characteristics of mitigating systems f
that are permitted by procedure to be inoperable for any period during operation shall be selected so that values that would result in violation of the acceptance criteria would r.ot be l
l expected to occur during
~
[(A3] most of the design lifetime of the facilitv[ies-issued eperating-iicenses-be'ere-danmary-i--1984 er ef-facilities
23 standardiced-te-a-facility-at-the same site-that-was-issued an eperating-iieense-before-deneary-1;-1984r]
[(B3-aimest ai?-of-the-design-iifetime of-facilities-issued-epera-ting-iicenses-en or-after-daneary-1;-19647-except-faciiities s tandardi z a d-te-a- f acii i ty-at-t h'e-s ame-s i t e-that-vss-i s s ue d an-eperating-iicense-befere-danmary-iT-1984r]
(3) ATWS mitigating systems shall be independent, separate and diverse from tiie reactor protection system.
ATWS mitigating systems shall be designed, qualified, monitored and periodically tested to assure continuing functional capability under the conditions 4
accompanying postulated ATWS' events including natural phenomena such as earthquakes, storis, tornadoes, and hurricanes, and floods expected te occur during the design life of the plant.
ATWS mitigating systems shall be automatically initiated when the conditions monitored reach predetermined levels and continue to perform their function without operator action unless it can be 1
demonstrated that an cperator would have adqquate information and would reasonably be expected within the time available to l
take the proper corrective action.
l l
(4) Evaluation models, as defined in paragraph (b) (2) of.this sec-tion, together with the description and results of the analysese and tests necessary to verify the validity c7 the assumptions 9
g% -
24 made in preparing such evaluation models, shall be submitted to the Nuclear Regulatory Commission for approval by March 1,1981 or prior to issuance of an operating license, whichever is later.
(5) A description of all measures to be taken to ensure compliance with the criteria set forth in paragraph b(1) of this section together with such proposed changes in technical specifications and license amendments as may be necessary to ensure compliance with such criteria sha?1 be submitted to the Nuclear Regulatory Commission as follows:
(i)
For all light-water-cooled nuclear power plar.ts for which
~
operating licenses have been issued on or before August 22, 1969, such inforaation shall be submitted no later than December 1, 1981.
(ii) For all light-water-cooled nuclear power plants for which operating licenses have been issued after August 22, 1969, such information shall be submitted no later than July 1, j
1981 or prior to issuance of an operating license, whichever is later.
(6) Those measures necessary to ensure compliance with the criteria set forth in paragraph (b)(1) of this. section sha?1 be imp'emented on'the following schedule:
25 (i)
For all boiling water reactor power plants for which operating licenses have been issued on or before August 22, 1969, all modifications shall be completed by Janaury 1,1984.
(ii) For all boiling water reactor power plants for which operating licenses have been or may be issued after August 22, 1969, all modifications shall be. completed no later than July 1, 1932 or prior to issuance of an operating license, whichever is later.
(iii)
For all pressurized water reactor power plants, all radi-fications shall be completed no later than January 1, 1984 or prior to issuance of an operating license, whichever is later.
1 (c)
(1)
In addition to those requirements set forth in paragraph (b) of j --
this section, each light-water-cooled nuclear power plant. except as provided in paragraph (c)(2) of this section, shall be p.rcvided with:
l (i) Actuation circuitry for ATW5 mitigatinp systems that is I
independent, and diverse from the reactor protection i
system; and-m O
r r
E
26 (ii) Prompt automatic containment isolation initiated by a significant source af radiation in the containment resulting from failure of the fuel rods following postulated ATWS events; and (iii) Those modifications necessary to reduce the common mode failure potential of the control rod scram discharge volume in plants designed by the General Elettric Company including diverse scram discharge volume level sensing devices; and-(2)
In addition to those requirements set forth in paragraohs (b) and c(1) of this section, each pressurizec' light-water cooled nuclear power plant designed by the Babcock and Wilcox Company or the Combustion Engineering Company shall be provided with
[(iv3] Those modifications necessary to provide a supplementary l
reactor trip system that is diverse from the reactor trip portion of the current reactor protection system, l
(2) Presserized-iight water-cecied n ciear power piants-issued eperating-iicenses-befere-dancery-i--1984-er standardized-to-a l
faciiity at-the same-site-that wr.s-issued-an-eperating-iicense before-dancery-1--1984-need net coNpiy-with-the-requirements ef 1
/
27 paragraph-(c3(13(iv3-if-the-facility-cenferes-to-the requirements ef paragraph-(b3 ef-this sectien except-that-the-fraction of-the design-iifetime esed-te-determine-the-vaice of parameters shaii be greater-than-that-specified-in paragraph-(b)(f)(i3-(3) A description of the measures together with such proposed changes in technical specifications or license amendments as may be neces-sary to ensure compliance with the criteria set forth in paragraphs (c)(1), and (c)(2) shall be submitted to the Nuclear Regulatory Commission no later than July 1, 1981 or prior to issuance of an operating license, whichever is later.
(4) Those measures required under paragraphs (c)(1) and (c)(2) of this section shall be completed by July 1, 1982 or prior to issuance of an operating license, whichever is later.
All interested persons who desire to submit written ccaments or suggestion con-cerning the proposed rulemaking should send their comments to the Secretary of the Commission, U. S. Nuclear Regulatory Commkssion, Washington, D. C.
- 20555, Attention:
Docketing and Service Branch, or on before Copies O
e G
28 of comments received on the proposed amendments may'be examined in the Commission's Public Document Room at 1717 H Street, NW, Washington, D.C.
(Sec. 161b and i, Pub. Law 83-703, 68 Stat. 948, Sec. 201, Pub. Law 93-438, 88 Stat. 1242 (42 U.S.C. 2201(b), 5841).)
Dated at this day of
, 1980__,
For the Nuclear Regulatory Commission.
Samuel C. Chilk Vecretary of the Commission
- 901 days after publication in the Federal Register O
4 3
m
i i
1 J
4 f
F r
t I
ENCLOSURE L t
e O
4 O
b v
e.y-m_.-,r-,
e'-----.------,e-._
/
PROPOSED ATWS RULE ACCEPTANCE CRITERION PRIMARY SYSTEM PRESSURE I
The specification of an allowable pressure following an ATWS event in the acceptance criteria of the proposed rule should be based primarily on assuring the integrity of the reactor coolant system.
However, the degree of assurance is difficult to quantify since the probability of failure of the reactor coolant system as function of stresa level is poorly defined.
Therefore, the specification of a pressure limit must be based almost entirely on judgment. The ASME Boiler and Pressure Vessel Code Service Limits are defined in terms of consequences and not in tems of probability of failure.
The Level C Service Limits " permit large defomations in areas of structural discontinuity," whereas the Level D Service Limits "pemit gross general deformations with some consequent loss of dimensional stability..." The Level C criterion limits the primary membrane stress due to pressure and other mechanical loads to the yield strength of all materials.
In addition, for ferritic materials, the primary membrane stress due to pressure only is limited to 90 percent of the yield strength.
The level D criterion limits the primary membrane stress to about 70 percent of the ultimate strength of materials. The internal pressures that would result in reaching the Level C limit would be approximately 20 percent to 30 percent greater than the design pressure and the pressure at the Level D limit would be another 20 percent to 30 percent greater than the pressure at the Level C limit.
The ASME Code does specify overpressure limits for normal, upset, and emergency conditions while leaving to the designer the detemination of the specific events to be included in each category.
Overpressure for upset conditions is limited te 10 percent ~above the Design Pressure.
In 1978, the Code,eas revised to include a specific requirement for over-pressure protection of ASME Class 1 components for emergency conditions.
The Code now limits stresses in emergency conditions to those corresponding to the Service Limit C.
The Code does not provide
-rules for overpressure for faulted conditions.
None are provided because the Level D Service Limit was never intended to apply to situations where, as in ATWS events, the major portion of the load results from pressure within the component.
Because of this, the Level D allowable stresses for materials such as those used for bolts were not developed and are not included in the Code.
In current NRC practice, the correlation between the conservatism of acceptance criteria and the probEbility of an event is not uniform.
For anticipated transients, the Level B limit is applied and overpressure is limited to approximately 10 perednt greater than design.
Other events of lower probability do not result in exceeding the design pressure of the reactor cooltnt system.
However, steam generator tubes are subjected r
to approximately the same differential pressure during a steamline-break accident as during an ATWS event.
Integrity of the tubes following a steamline break has been found acceptable based on extensive testing of tubes with. defects.
Since. tubes are subjected to stresses in the plastic range, use of the test data to demonstrate integrity of the tubes is
[
equivalent to applying the Level D Service Limit to these components.
L-1 u
i Although a directly applicable precedent is not available, the criteria applied to other components and structures provide some comparison. The acceptance criterion for fuel is that fuel damage is limited to some
~
clad perforations following anticipated transients.
Greater fuel damage, i.e., severe oxidation and ballooning, is permitted following a loss-of-ccolant accident. The acceptance criterion for the containment is mo're restrictive. While there is no containment pressure acceptance criterien for anticipated transierits, the containment pressure following a loss-of-coolant accident is limited to the design pressure, which is analogous to the Level A Service Limit for the reactor coolant system.
The conservatism of an acceptance criterion is not only a function of the criterion itself, but also of the conservati'sm of the evaluation model used to calculate the parameter being considered. Thus, changes in the values of the parameters used in the evaluation model, such as the moderator temperature coefficient, can affect the conservatism of the requirement as much or morg so than changes in the value of the acceptance criterion.
If adopted, the Level D Service Limit would not be the limiting factor for overpressure following ATWS events. The additional requirement to demonstrate the operability of components necessary for shutdown following an ATWS event would limit the pressure to below that which would result in Level D stresses.
In some cases, the demonstration of operability may be a limiting factor even at the Level C limit.
A specific demonstration of the capability of the safety and relief valves to close after discharging water at pressures corresponding to the Level C limit has not been
~
required. Some extrapolation of test data at low pressures and engineering jud;:ent have been used to justify that these valves will close following ATW5 conditions. The industry now has underway a program to test the operability of safety and relief valves discharging water at design pressures.
Operability of these valves at pressures corresponding to the Level C limit would still require extrapolation from the test conditions. Operability at pressures corresponding to the Level D limit would require even further extrapolation.
Bolted connections, such as the vessel head and steam generator manways, would not be expected to maintain the pressure integrity of the connection if subjected to a pressure corresponding to the Level D limit.
Analyses performed by the NSSS vendors and an independent analysis performed by INEL for the NRC indicate that the vessel closure seal would leak at a pressure somewhere between 3400 and 3S00 psi, depenoing on the specific vessel design.
Limiting stresses to the Service Limit C provides reasonable assurance that gross coolant leakage would not occur.
The results of the preliminary fracture toughness analyses indicate that, for PWR vessels fabricated from materials with a relatively low " upper shelf" or high temperature toughness and more susceptible to radiation damage, brittle fracture of the vessel could be of concern at pressures below those corresponding to Service Limit D, assuming the presence of an AS!'.E Code reference flaw.
At pressures below those corresponding to Service Limit C brittle fracture would not be a concern.
6 L-2
a.
r_
In addition to technical arguments concerning the adequacy of the ATWS acceptance criteria, the staff has considered the amount of analysis by the industry and review by the staff necessary to cemonstrate compliance with the criteria.
Beyond calculating the system pressure during postulated ATWS events, little or no analysis has been required to demonstrate that stresses are limited to the Service Limit C.
In most cases, the value of the already calculated normal operating or upset condition stress increased by the ratio of the calculated ATWS pressure to the design or upset condition pressure has been accepted. Adoption of the Service' Limit D would recuire the licensees and applicants to perform elastic-inelastic stress analyses of the reactor pressure boundary which are more complex than the. elastic analyses n.eeded to calculate Service Limit C stresses.
The proposed criterion allows a demonstration of integrity to be based on test data, so that a few components, such as the steam gene,rator tub'es, would not unnecessarily ~be the limiting fact'or, in determining the acceptable pressure.
However, this alternative method would be limited to local areas in the reactor system.
The extensive deformations that would be calculated to occur at the Service Limit D would result in acments not present at the Service Limit C pressure levels thus increasing the complexity of the calculatter.s further.
Adopting the Level D limit over the Level C limit would increase the cost of analyses to the licensees and applicants and increase the staff review effort with a modest, if any, increase in the allowable pressure.
The specification of an appropriate.:.lowable pressure following an ATWS event has been an issue during the review of proposed requirements for ATWS events.
The arguments discussed here have been adcressed earlier and are presented in NUREG-0460 (Vol I Section 7.1.2 and Vol 3 Appendix D, Section G.1).
The staff, the ACRS and the ASME Code committees have been involved in evaluating the issue.
The consensus of these various groups is that the Snevice Limit C is the appropriate limit for ATWS events.
6 e
f e
e L-3 4
I Y
s i
i
}
i-I EtiCLOSURE M t
I I
9 4
I I
i r
6
PROPOSED ATW: RULE CONCEPTUAL DESIGNS FOR ADDITIONAL RELIEF CAPACITY.
If adopted, the proposed rule could require that additional relief capacity be installed on some PWR's. The staff has not made a design study to determine how additional valves could be installed on any particular plant, but has identified some conceptual designs. Some plants have extra nozzles on the pressurizer safety valve header.
It may also be possible to remove existing valves and. replace them with valves of larger capacity.
In scme cases, this may only recoire the modification of the valve internals to install a larger orifice. Since the additional valves would have ahigher set pressure than the existing valves and would pass water, it is not essential that they be connected to the pressurizer steam space.
Other possible locations are piping connections to the reactor coolant system such as the Residual Heat Removal letdown line, or the steam generator manway cover. The reactor vessel head vents are also possible locations, but because they are generally about one inch lines they ray be inadequate.
If protection of operating personnel from inadvertent opening could be assured, dit-charge piping to a quench tank would not be required.
e O
e d
M-1
(
s ENCLOSURE N f
C n
1 3'
I
.