ML20035B395
| ML20035B395 | |
| Person / Time | |
|---|---|
| Issue date: | 03/24/1993 |
| From: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Shewmon P Advisory Committee on Reactor Safeguards |
| Shared Package | |
| ML20035B396 | List: |
| References | |
| PROJECT-672A, PROJECT-674A, PROJECT-679A, PROJECT-680A ACRS-GENERAL, NUDOCS 9304010285 | |
| Download: ML20035B395 (9) | |
Text
_ _.
March 24, 1993 Mr. Paul G. Shewmon, Chairman Advisory Committee on Reactor Safeguards U.S. Nuclear Regulatory Commission j
Washington, D.C. 20555 l
Dear Mr. Shewmon-I am responding to the Advisory Committee on Reactor Safeguards (ACRS) letter of February 19, 1993, to the Chairman concerning the issues pertaining to the l
advanced reactor (PRISM, MHTGR, and PIUS) and CANDU 3 designs and their relationship to current regulatory requirements. The responses in the enclosure are organized into General and Specific categories, as are the comments in the ACRS letter.
I have commented upon those issues where the i
staff's position is questioned or where I felt our position needed some clarification.
The staff will continue to keep the ACRS informed of any changes in staff or
{
preapplicant positions, as well as any significant developments in the implementation of the policies.
l Sincerely, Origtialsigned by l
James IlTaylor James M. Taylor i
Executive Director i
for Operations l
'i
Enclosure:
Staff Response to ACRS Comments cc w/ enclosure:
The Chairman Commissioner Rogers Commissioner Curtiss 9304010285 930324 h
Commissioner Remick Commissioner de Planque PDR ACRS SECY GENERAL PDR I
Distribution:
MTaylor JGPartlow NRR Mail Room l
Central File (w/ incoming)
JKnubel HPastis 12/G/18 Project Nos. 672, 674, RMBernero PDAR R/F (ED0 #0008601) 679, and 680 ESBeckjord DMCrutchfield BToms PDR ELJordan MJCase PMagnanelli EDO #0008601 JFScinto RCPierson PShea JMTaylor TEMurley OGC PDAR Ticket JHSniezek 3100*?q FJMiraglia OPA File HLThompson WTRussell OCA THCox i
JBlaha ATGody, Sr.
EDThrom h@R OSee previous concurrence A)A D:PD JAR TECH EDITOR *
. JRR#
LA:PDST:A p SC:PDAR:ADAR SC PShea EDThrom #
TH
- sa RPie RSanders DCrutchfield y/f'/93 3/$f 3 //B/93 g/g93
$ /d 03/10/93
/
TEMrNe
$a J Ud..
- bY
.f.u h //]- f k/[8 tl f
RECORD Document Name:
ED008601 g
STAFF RESPONSE TO ACRS COMMENTS
GENERAL COMMENT
S j
Comment 1: We find that the identified issues are important and that the staff should receive guidance from the Comission.
(There are other policy issues affecting these reactor designs.that are being addressed in connection with the evolutionary and passive LWR designs.) There may well be additional policy issues that appear during the preapplication review process. The staff has comitted to identify any such issues in subsequent Comission papers.
Response
The staff will identify any policy issues that arise subsequently.
Coment 2: The staff has grouped these ten issues into the two categories i
described above. We note that all of the affected preapplicants who appeared before us would treat Issue I (Control Room and Remote Shutdown Area Design) as a Category 1 issue, whereas the j
staff proposes it as a Category 2 issue. We will discuss this difference of opinion below in our opinion on Issue I.
Response
No response to this coment.
t Coment 3: For Category 1 issues, the staff proposes more conservative i
alternatives than the preapplicants propose, in order to account
{
for uncertainties associated with the conceptual design.
We are concerned that such an approach might well freeze an unnecessarily
(
large degree of conservatism into the designs, and the preap-l plicants would have great difficulty persuading the staff to relax i
this conservatism on the basis of more precise information l
available in the final design.
Response
The staff's positions on Category 1 issues, with the exception of containment and emergency planning (EP), are not more conservative than positions proposed by preapplicants. The containment recom-l mendation is based on, and will be further developed from, the i
current Advanced Notice of Proposed Rulemaking issued by the Commission, and the staff's EP recomendation notes that further evaluation of the advanced reactor designs may permit some relaxation from current requirements.
The staff is not prepared to, nor does it intend to, specify firm standard design certification requirements at the preapplication review stage. The preapplication review approach is to examine a range of postulated operational events and accidents, including j
some accident sequences postulated to result in either substantial core damage alone or core damage with a large release from con-tainment. Such examinations are consistent with the direction of the Comission's Severe Accident Policy Statement, and with the ENCLOSURE
10 CFR Part 52 requirement to confirm, through a full-plant, design-specific probabilistic risk assessment (PRA), the margins available in the design to accommodate events of low probability.
These examinations will assist in specifying which accident sequences should be assigned to the frequency categories identi-fied as anticipated operational events, anticipated transients and accidents, and severe accidents.
It is intended that licensing r
basis consequence limits for accident sequences will be entirely consistent with the Commission's safety goal and severe accident i
policies.
For the preapplication review, the staff intends to develop a set
.of event categories for accident selection. These categories
't would be used for defining acceptable analyses criteria and consequences (dose limits) based on'the likelihood and potential significance of an event.
These categories would both encompass traditional events at light-water reactors (LWRs) and would extend to severe accidents. For anticipated operational transients, analyses assumptions would be conservative. For anticipated or design base accidents, the single-failure criterion would be used in conjunction with such conservative analyses assumptions as no operator credit and no use of non-safety-related equipment to mitigate the outcome of an event.
For severe accidents, a best-estimate analyses assumption would be acceptable as would justifiable operator actions and credit for non-safety-related equipment.
The analyses would, therefore, prescribe the event scenario, identifying the amount, timing, and magnitude of fuel and core damage, and the status of the integrity of the primary coolant boundary. This fuel and core damage would then be used to evaluate the amount and type of radionuclide releases from the fuel into the primary coolant. Uncertainties in the specific amount of radionuclide species released from the fuel would be assessed for the scenario but would not be further increased to provide some measure of additional conservatism. The status of the primary coolant boundary would then determine the releases from the primary coolant boundary into the containment (or confinement) region, and the containment performance would then determine the releases to the environment. With regard to conservatism, the effectiveness of systems and barriers to mitigate releases would be treated consistent with the appropriate event category.
In addition to using PRA results to identify an appropriate event category for some scenarios, the staff may, based on engineering judgment, require the preapplicant to evaluate some event sequences applying appropriate conservatism to, for example, the anticipated accident category rather than the severe accident category. The staff would need this assessment in order to assess margins in the design, to account for uncertainties in the PRA t
data base, and to ensure that the importance of systems needed to mitigate accidents is understood.
Coment 4: We support the staff recomendation that "a prototype CANDU 3 is not required for design certification."
Response
The staff and ACRS agree on the staff recomendation.
Coment 5: We support the staff intention to notify the Comission if its position on any of these ten issues should change, or if new issues are identified.
Response
The staff and ACRS agree on the staff recomendation.
Coment 6: We have no objection to the staff recomendation that the highest priority be given to issues that are applicable to the PRISM design.
Response
The staff and ACRS agree on the staff recomendation.
Coment 7: We understand and sympathize with the staff recommendation to defer decisions on generic rulemaking on these ten issues.
Nevertheless, we urge the Commission to address these decisions in the near future.
(The generic rulemaking question may arise in connection with passive LWR designs.)
Response
The staff and ACRS agree on the staff recomendation.
Coment 8:
In several places in the draft Commission paper, there occurs qualitative language, e.g., " appropriate conservatisms" or
" credible severe accidents." This language must ultimately be translated into quantitative guidance. We believe that the quantitative guidance is, to a large measure, policymaking, and should not be relegated to low-level reviewers.
Response
The staff work on the advanced designs has not progressed to where it can quantitatively define " appropriate conservatism". or
" credible severe accidents." We agree that such definitions i
should be approved at appropriate agency management levels, and will not be relegated to " low-level reviewers." The staff will delete the words " appropriate" and " credible" from the policy issues paper, as suggested by the ACRS.
I SPECIFIC COMMENTS Category 1 Issues Comment:
A.
Accident Evaluation t
The staff proposal to develop a single approach with certain specified characteristics appears reasonable. We would like to review that approach when it is ready.
We believe, however, that
, l
4 the staff should identify at an early stage quantitative guide-lines and criteria for accident selection and evaluation. We note that AECLT has taken exception to some of the statements in the draft Comission paper that relate to its approach to this issue.
We believe that this disagreement can be resolved by AECLT and the staff.
Response
The staff plans to propose, as soon as possible, quantitative guidelines for design certification accident selection and evaluation. This will include definitions of frequency categories and associated acceptable consequence limitations assigned to categories and specified accident sequences.
Coment:
B.
Source Term The staff proposal to base the source terms on mechanistic analyses appears reasonable, although it is clear that the present data base will need to be expanded. We note that the staff is now developing for LWRs a revision to the TID-14844 source term.
It will be appropriate for the staff to consider using the newer approach when it develops source terms, and to take specific account of the unique features of each of the reactor types.
Response
The staff and ACRS agree on the staff recommendation.
Comment:
C.
Containment The staff proposal "to postulate a core damage accident as a containment challenge..." appears reasonable. We would like to review the list of postulated accidents when it is ready.
Response
The staff and ACRS agree on the staff recommendation.
Coment:
D.
Emergency Planning The staff proposes that advanced reactor licensees be required to develop offsite emergency plans which will include a requirement for onsite and offsite exercises. This proposal appears reason-able under the present circumstances, except that we would follow existing LWR guidance that permits the omission of offsite exercises when it can be shown that the design would preclude any accidental release exceeding the EPA Protective Action Guides.
The staff has agreed to consider, after a review of Accident Evaluation (Issue A, above), whether some relaxation from current requirements may be appropriate. We urge that work on Issue D be closely correlated with work on Issues A and B, in order to avoid unnecessary conservatism.
Response
The staff and ACRS agree on the staff recommendation.
, i
Coment:
E.
Reactivity Control System The staff proposal that the absence of control rods need not disqualify a reactor design, provided that an applicant can show a level of safety in reactor control equivalent to that of a traditional rodded system, appears reasonable. We note that this issue is applicable only to the PIUS concept, and that we have not yet had the benefit of presentations by the PIUS designers.
Response
The staff and ACRS agree on the staff recomendation.
Coment:
F.
Operator Staffing and function The staff intends to review the justification for a smaller crew size by evaluating the function and task analyses for normal operation and accident management. This intention appears reason-able, although we~ believe that particular attention needs to be given to multiple module designs. We note that this issue is related to a similar issue for passive reactors. We believe that the Commission policy should be the same for the advanced reactors and CANDU 3 as it is for the passive reactors.
Response
The staff and ACRS agree on the staff recommendation.
Coment:
G.
Residual Heat Removal The staff belief that reliance on a single, completely passive, safety-related residual heat removal (RHR) system may be accept-able appears reasonable, although we would have liked to see the criteria to be used by the staff in deciding acceptability. We agree with the staff that NRC regulatory treatment of non-safety-related backup RHR systems for these reactors should be consistent with design requirements (not yet identified) for passive LWRs.
Response
The staff and ACRS agree on the staff recommendation.
Coment:
H.
Positive Void Reactivity Coefficient i
We agree with the staff that the existence of a positive void reactivity coefficient is a significant concern, but that it should not necessarily disqualify a reactor design. The burden of showing that the consequences of those accidents that would be aggravated by a positive void reactivity coefficient are either acceptable or could be satisfactorily mitigated by other design features surely falls on the preapplicant. On the other hand, the staff should state the criteria it will use to judge " acceptable" or " satisfactorily."
Response
The staff and ACRS agree on the staff recommendation. The staff will define the criteria to judge " acceptable" and
" satisfactorily" as it performs the preapplication review and during this review will develop an understanding of the behavior,
i
[
of the particular design during positive reactivity insertion j
transients and accidents.
l Category 2 Issues Coment:
I.
Control Room and Remote Shutdown Area Design We do not agree with the staff decision to treat this issue as a I
Category 2 issue, and the concomitant recommendation to apply current LWR regulations and guidance until passive LWR policy in I
this area is finalized. We believe that this issue should be a j
Category 1 iue, and that the preapplicants should accept the i
burden of convincing the staff that a proposed-design is satis-l factory, according to some criteria that should be specified by
}
the staff.
Response
The staff is recomending a policy position to be utilized at the l
preapplication review stage. As stated in response 3 (above), the i
staff does not intend to shut off further dialogue with preap-l plicantr, but to state that, at this time, justification for j
recomending departure from current requirements is not estab-e lished. As noted in the policy paper recomendation, this issue l
siso is dependent on further development within the context of l
advanced passive plant design reviews. The staff believes that preapplication design review should be completed by identifying
- and. evaluating differences between preapplicant proposals' and.
current LWR requirements,' before recommending design certification l
requirements.
l The staff believes that its current recomendation in the policy i
paper remains appropriate, that is, to evaluate preapplication i
designs by comparison to current LWR requirements / guidance for the y
~
imediate future.
As advanced passive LWR policy.for design requirements;of control-rooms and remote shutdown facilities is i
developed, and~ as further evaluation of the completely passive
~
shutdown and decay heat removal functions in the Department of Energy designs are further evaluated, additional changes to relax requirements on main control rooms may be justified for the designs treated in this paper.
i Coment:
J.
Safety Classification of Systems, Structures, and Components
.i (SSCs)
This issue is relevant only to the MHTGR concept. GA makes a l
I persuasive case that the MHTGR is sufficiently different that the t
LWR criteria for. identification of safety-related structures, systems, and components should not arbitrarily be applied to the j
MHTGR. We concur with this view and believe that Issue J should also be classified as a Category 1 issue. This would not preclude coordination of the policy for passive reactors with the policy j
for the MHTGR.
> i i
Response
The staff believes that this issue is fundamental to the main-tenance of the " defense in depth" philosophy and policy held by the Commission over many years of LWR regulation. The Nuclear Regulatory Commission (NRC) has long held it important to protect the integrity of the reactor coolant pressure boundary, to protect the SSCs necessary to shut down the reactor and maintain decay heat removal, and to protect the SSCs needed to prevent or mitigate accidents and minimize releases.
This policy is believed sound and necessary to the partitioning of the risk of radioactive releases among several independent barriers to radionuclide transport. The approach taken by the preapplicant for the modular high-temperature gas-cooled reactor (MHTGR) would limit NRC regulation to requiring safety-grade classification only on SSCs needed for preventing or mitigating offsite releases. The staff feels that this approach could result in placing reliance on too few SSCs, therefore, not assuring multiple, highly reliable barriers to radionuclide release.
This issue deals only with determining which SSCs should be classified as safety related, making it equivalent to the specification of SSCs as " safety grade." For advanced reactor designs, the classification as safety grade may not necessarily mean that such SSC should be designed, built, tested, and maintained to a single set of standards that includes seismic Category I, electrical Class 1E, and other traditional " safety grade" attributes. That is, having specified which SSCs should carry the safety-grade classification, consideration should be given, in the language of General Design Criterion 1, to design, fabrication, erection, and testing to quality standards commensurate with the importance of the safety functions to be performed.
6 go.a uang g
jo UNITED STATES g
8* Tnt NUCLEAR REGULATORY COMMISSION n
l i
a,E
('j 'D p py p.,
I WASHWGTON. D. C. 20555
'++
EDO Principal Correspondence Control
[
FROM:
DUE: 03/17/93 EDO CONTROL: 0008601 DOC DT: 02/19/93 FINAL REPLY:
Paul Shewmon ACRS
/
/
./ -
TO:
i c o w cd_,.2 D.3 / 9 3 gQ L c9%
Chairman Selin i
FOR SIGNATURE sF:
- GRN CRC NO:
Executive Director DESC:
ROUTING:
ISSUES PERTAINING TO THE AD/ANCED REACTOR (PRISM, Taylor MHTGR, AND PIUS) AND CANDU 3 DESIGNS AND THEIR Sniezek RELATIONSHIP TO CURRENT REGULATORY REQUIREMENTS Thompson B1aha DATE: 02/22/93 Mat Taylor Knubel ASSIGNED TO:
CONTACT:
Bernero, NMSS NRR Murley Beckjord, RES Jordan, AEOD SPECIAL INSTRUCTIONS OR REMARKS:
Scinto, OGC PREPARE RESPONSE TO ACRS FOR EDO'S SIGNATURE.
PUT COMMISSIONER AND SECY ON CC (SHOWN ON ORIGINAL) FOR REPLY,
'fy[/C f$ & d?I-' rib 2/93 WE,c [icf%h ; D7"Sg?mi=A Ou'b W M ; & & vJ -
- ? L h e
~ ~ ~ ',. ~,,!
[9i9fd' /tc
- '~~AYN!
yQ4:
((
- p.,_
E :) Ja Jggn 9 l
f%)6w l,y G Q'Ma I
Q &s( W
^-
a-4;vhDW wr