ML20040E251
| ML20040E251 | |
| Person / Time | |
|---|---|
| Issue date: | 10/06/1981 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1905, NUDOCS 8202040074 | |
| Download: ML20040E251 (13) | |
Text
hhk-l-lV-Sb O
/Vl'AS- /Vo6 S
U NN DATE ISSUED: 10/6/81 w
U 4
12 a
MEETING MINUTES g
OCTOBER 2,1981 WASHINGTON, DC
Purpose:
The purpose of the meeting was to discuss the ATWS Rule recently proposed by the Commission for public comment. Discussion focused on the portion of the Rule that incorporates risk assessment methodolgy.
Attendees:
Principal attendees of the meeting are noted below:
ACRS NRC W. Kerr, Chairman F. Rowsome J. Ebersole, Member W. Minners J. Ray, Member D. Ward, Member Utility Group on ATWS S. Ditto, Consultant E. Burns (SAI)
E. Epler, Consultant D. Edwards (Yankee Atomic)
S. Saunders, Consultant P. Boehnert, Staff *
- Designated Federal Employee l
A complete list of attendees is attached to the office copy of these minutes.
l Meeting Highlights, Agreements, and Requests 1.
Dr. F. Rowsome (RES-DSR) discussed the alternate ATWS Rule that he ghosted l
at the behest of former NRC Chainnan J. Hendrie.
Dr. Hendrie felt an alter-nate rule to the Staf f's, proposal was needed to give the rule sharper focus on deficiencies warranting backfit, and strengthen the incentives working on licensees to assure safe design-and operation. Also there was concern that l
the Severe Accident Rule may render moot the ATWS fixes mandated by the Staff S$
rul e.
l ujo 5
The heart of the alternate rule is the establishment of a reliability assurance program designed to find and correct reliability deficiencies in
~o y
lE" systems that prevent or mitigate ATWS events.
In addition, some hardware l
fixes are required to allow the opportunity to learn from experience with ATWS precursor events without suffering large radiation releases.
?
ATWS Mtg 10/2/81 The. reliability assurance program would require:
(1) ATWS accident sequence analysis, (2) operator training and instrument adequacy for accident diagnosis and mitigation, (3) cataloging of blindspots in testing and service experience where reliability faults may escape detection and repair, (4) study of common cause failure susceptibility in critical systems, and cost-effective improvements, and (5) maintenance of a continuing reliability assurance program.
Dr. Kerr felt Item (3) should be applied to plant safety generally.
Dr. Rowsome agreed, stating his belief this should be a USI item applicable to all accident types.
In response to a question from Dr. Kerr, Dr. Rowsome said he believes this Program is workable in a finite time frame.
The mitigation provisions of the Rule include:
'Large Modern BWRs 1.
Recirculation pump trip.
2.
Automated and upgraded liquid poison system as required so HPCI or l
RCIC can handle core cooling during plant recovery.
3.
Reliable scram discharge.
i l
'Large Modern PWRs_
1.
Autostart for auxiliary feedwater (already installed).
2.
Qualify instruments required for diagnois and recovery from ATWS.
3.
Qualify valves required to function to recover from ATWS.
j l
~
'All Plants 1.
Develop LCOs to minimize exposure to operation under conditions that would compromise ATWS mitigation.
2.
Mitigation criteria defined as: coolable geometry in core, containment pressure and temperature in design envelope, and no acute fatalities or l
substantial offiste property damage.
i
ATWS Mtg 10/2/81 Subcommittee discussion ensued.
Dr. Rowsome said the BWR requirement for both HPCI and RCIC availability is designed to try and intercept ADS and prevent suppression pool heat-up problems.
In response to Mr. Ebersole's questions, Dr. Rowsome said the acceptable damage threshold in his Rule is more severe than Part 100 limits.
A TMI-outcome would be acceptable under this Rule. Mr. Ditto said a recent EPRI study of operator reliability showed that the operators didn't always follow their own procedures. Mr.
Ditto doubted that the operators would quickly activate liquid poison in the event of an ATWS.
Mr. Ward commented that he felt the Rule was too generalized and too vague - it only exhorts licensees to go out and "do better" and NRC would tell them if that's OK or not.
In a response to Mr. Ebersole, Dr. Rowsome said he believes the present regulatory system provides large disensentives to safety.
His goal is to reward some of the most capable people associated with reactor safety (utility people) for doing their job well. He said the utility's investment in the plant is hostage to public health and safety.
2.
Mr. D. Edwards (Yankee Atomic) provided some opening comments for the Utility Group on ATWS (Group). The Group, consisting of 22 utilities, was formed after issuance of the last volume (Volume 4) of the NRCs NUREG on resolution of ATWS (NUREG-0460). Mr. Edwards listed four central concerns of the groups:
(1) NRC has had a "helter-shelter" change of ATWS solution approach through Volumes 1-4 of 0460; (2) the solution was continually growing; (3) the utilities could see a large resource drain for ATWS resolution vis-a-vis other more pressing requirements (TMI, etc.),
and (4) the ATWS fixes in. Volume 4 of 0460 had downside consequences to plant safety.
The Group will commit to certain modifications to its Members' plants (Figure 1) providing these fixes end the problem.
In response to a question from Mr. Ray, Mr. Edwards estimated that the Groups's proposed fixes should reduce the preventative ATWS risk by a factor of 2 to 4.
Mr. Ward objected to a statement by Mr. Edwards that the proposed fixes
ATWS Mtg 10/2/81 would reduce ATWS risk to the value stated in the ACRS Safety Goal report (NURE-0739).
Mr. Ward said the value in 0739 represents an approach not a recommended " safety number". As a result of further discussion, Dr. Kerr urged that the NRC and the Utility Group enter into discussion on possible solutions to the ATWS issue.
3.
Mr. E. Burns (SAI) discussed the results of an SAI study that made use of probabilistic methodology to demonstrate the relative improvement in safety obtained by implementing the Group's proposed ATWS fixes. The study:
(1) focused on relative safety improvement, (2) used operating experience to the extent possible, (3) treated PWRs and BWRs separately, and (4) used NRC estimates (e.g. scram system failure rates) for parameters where there was an unresolved conflict between the Group and the Staff.
The results (Figure 1) were stated to appear consistent with other PRA work.
ATWS is shown not to be a significant risk contributor, while for BWRs, ATWS is one of several existing risk contributors.
In response to questions from Dr. Kerr and Mr. Ray, Mr. Burns,said he would estimate the PWR ATWS risk contribution to be on the order of N5%.
For BWRs, ATWS is one of 3 or 4 dominant risk contributors.
In a summary statement, Mr. Burns said the study leads to the following conclusions for BWRs - implement the Utility Group recommendation immediately and await further study and an action plan on an integrated approach to degraded core response before requiring an automated liquid poison system.
For PWRs, the Group's fixes should also be implemented with the comment that further reduction in an already low frequency of unacceptable plant conditions is difficult to justify.
Mr. Burns detailed the methodology used in the study for a typical PWR and BWR. The assumptions used in the analyis and an example of typical event tree for each reactor type were shown (Figures 2-5).
There were also a number of conservatisms listed for the evaluation (Figure 6).
l ATWS Mtg 10/2/81 4.
Mr. W. Minners (NRC-DST) provided comments on the Hendrie Rule proposal. He expressed support for the plant reliability assurance program as well as the proposed requirement for periodic testing and maintenance to eliminate "blindspots" in equipment reliability. His chief concerns with the Hendrie proposal are:
the lack of specificty on what is an acceptable degree of conservatism for meeting the acceptance criteria in the Rule; and (2) the Rule does not contain consistent requirements between PWRs anc BWRs (too lenient of PWRs) and this is not a proper bais for regulation. Other concerns noted include the need for evaluation models (lacking in Hendrie Rule), the treiitment of pre-1969 plants (Rule puts the burden of proof for safe operation on NRC, not the utility), and the possibility of overly restricting plant operation with the use of LCOs.
5.
In open executive session, Dr. Kerr solicited Subcommittee comment for input to his report to the full Committee scheduled for the October 1981 meeting.
'J. Ebersole - Doesn't believe the PRA studies will result in plants meeting the safety goals he'd be comfortable with. Wants to see pubic comments on the Rule.
'J. Ray - Endorses the Hendrie approach. Relieves NRC has been too precriptive in its required fixes. There's alot of talent in the ir;dustry that can be fruitfully applied in this area.
'D. Ward - While the Hendrie approach to regulation is interesting and should be pursued this approach is not appropriate for resolution of ATWS.
ATWS requires a quicker solution than afforded by this approach.
The Utility Group proposal doesn't go for enough - a factor of 2 reduction in ATWS risk vis-a-vis these fixes is not impressive.
l
'S. Ditto - Perfomance deficiencies are the major problem in reliability.
Systems need to be made more resistent to variations in operator performance.
He favors the NRR approach to ATWS resolution since he believes ATWS is a system problem rather than a procedure problem.
ATWS Mtg 10/2/81
'E. Epler - Hendrie proposal is long overdue but he feels the industry doesn't have the people available to do what the Hendrie Rule requires.
He feels the Industry should work to reduce the frequency of challenge to the scram system.
'S. Saunders - Supports the Hendrie approach and dislikes the prescriptive-fix approach of the Staff.
Dr. Kerr said he would take the Subcommittee comments under advisement.
The meeting was adjourned 1:00 p.m.
l l
NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room,1717 H Street, N.W.,
Washington, D.C., or can be purchased from Alderson Reporting Company, Inc., 400 Virginia Avenue, S.W., Washington, D.C. 20024, (202) 554-2345.
l t
yl.y J
UTILITY GROUP ON ATdS PETITION REQUIRDGTS i
?
=-
MODIFICAT10fi VDiDOR 1.'es tinghouse (a)
Automatic Auxiliary Feedwater Initiation independent of the RPS (b)
Automatic Turbine Trip independent of the RPS Babcock and Wilcox (a)
An alternate means to initiate control rod and injection that is diverse Combustion Engineering from, and redundant to the i
electrical portion of the RPS up to but not includ-ing the trip breakers.-
(b)
Automatic auxiliary feedwater initia-tion independent of the RPS.
General Electric (a)
Recirculation Pump Trip (b)
An independent, redundant and diverse electrical means to initiate a reactor scran upon receipt of a signal indicative of an AT'<JS (c)
Scram discharge volume modifications f/6.l
~ ~ ~.
LEGEND m Safety Goal Hazard State 1 Core Dan; age Freq. (Per Reactor Year.
i
-O-O--G-I NUREG 0460 BWRs 7
5 E
b 10-4 3
E s
il%
u/
m
!!/
3
- j L"
!l/
E
!!/
E
-5
- /
u 10 6
?/
jj Pre-Petition
- e g
v/
/
u h
Post-I
!f
!!/
E Petition ii/
- /
- l/
if
/
J:
i/
j
!h ih
.. /
j c/
- ?h vi v/
v/
l
/
10-0 W
PWRs flb - lb
i 4 BWR
(
ASSLIPTIONS t
i e
1.
A GENERIC BWR IS CHOSEtt FOR EVALUATION.
(SIMILAR TO A B W 4 SUCH
~
AS PERH BOTTOM) 2.
ARI EFFECTIVENESS I'AY BE lhDERSTATED, H WEVER, IT IS BASED 04 THE DISTRIBUTION OF HISTORlCAL FAILURES km PRECURSORS.
3.
ARI ktD RPT USE SEPAoATE he DIVERSE SENSORS k!D l.0GIC FROM THE RPS, EACH WITH RELATIVELY HIGH RELIABILITY.
f4.
OPERATOR ACTION IS 60'ERNED BY THE PRESCRIBED PROCED'JRE.
5.
OPERATOR TRAINING EMPHASIZES THE NEED FOR klARENESS OF POTENTIAL ATWS EVENT.
6.
SLC FAILURE IS lbMINATED BY OPERATOR FAILURE TO INITIATE PROMPTLY RATHER THAN thRD.1ARE FAILURE.
SWAIN CUTTt%N IS USED AS BASIS FOR O'JANTIFICATION.
7.
!'ANuAL INITIATION OF RHR WITHIN 15 TO 30 Mit:UTES IS OUANTIFIED USING SWAIN GUTTt'AN EVALUATION OF PWR AUXILIARY FEED.1ATER I'ANUAL INITIATION.
laittator RPS RP5
- sectre, adeouste Safety Coolsat Medvertent Operettoa_
Meat Float I
(
- 3 ' "*
milf
'echaal(gl (Ie({r$Cgt Pymp A%l Preggurg V31pel lajettton Adt act FW or HPCI Eraeval Sequaates Coadttfea C
& Manwal Cloture fete Control Rectose Actuated 00 Not Rua s.,,
Fea 64ter Post level Con t rol
'8' g
a E
M E 2 T,2 g
gg i
lWI II (E pv KP n
l 10' Ii t'IU KP n
M
't 10' II
'O KP I
n
(
g g jy 108 I !j i
!10'#
,I{ [r i t *j KP M
'P CE
[10
'M i
10'
'nI ('
gp 10' Il (El PO gp
'y g
'pl (E}PU LOI I10'#
f2 a PUO n g g,
r,r, pC,
,,C,
7:10-5 10'
' ')
t e
0 MP t"
Id u
') M
,9 E UI HCP i,r t xA. KP 10*8 r,2 C(4 g,
It C n
?e Ot I
n'i f 2 C, W gp n
g,2 f ? C,U, gp g
i 10-
'M M0 KP t,7 C, U
LO!
.08 l
2 fiC UD i
g m gp l
'ni C, P bt d
I 10'#
Y2C g g pu uc,
-2 l
10 Y 2 C, N g
g ncy in-3
'nr C,,o g,
r r?C PU J,
.08 4
LOI 1
-2 17 2'T h0 MP 1
I Y l C
g y
gp l
)
8 IIhU M
p KP
(
^
'nI m 2 LOI
,g I
.9
[
M M 2 UO pgp 4
l I IE M M M
. 0CA. KP 10 3 M
t N Based upon high power taittators caly.
Event Tree Diagram For Postulated ATWS Accident Sequences Following A M51V Closure From High Power.
,./
PWR EVENT TREE EVALUATION ASSUf1PTIONS
{XAMPLECHOSENFORIHISPRESENTATIONISAGENERICWEsIINGHOUSE e
PLANT e
EXCEEDANCE OF A PREDETERMINED PRIMARY SYSTEM PRESSURE REPRESENTS CONDITIONS BEYOND IHE GUIDELINE FOR IHIS STUDY e
MODERATOR IEMPERATURE COEFFICIENT IS A KEY VARIABLE IN PWR EVALUATION.
MODERATOR IEMPERATURE COEFFICIENT DIVIDES IHE RESPONSE SURFACE OF PWRS INTO REGIMES OF ACCEPTABLE AND UNACCEPTABLE PLANT CONDITIONS FOLLOWING A POSTULATED ATWS GIVEN IHAT OTHER SAFETY SYSTEMS RESPOND AS DESIGNED.
e ADDITIONAL SYSTEM FAILURES'WILL RESULT IN HIGH PEAK PRIMARY SYSTEM PRESSURES N
y
__ s 1
~
Y I
...........aaan:
.....a=........
w
_a
_a Y
- a mia$$$$$ fit.-s$$$tst E
1 I
3 6
. T. J T. $ 5 T. I. T. i. t. *a *4
%. %. I. I. T. } '. t. *51 *5. 4 14. ~&. '.
'm E_
g
- :::::::::.~ ; ; ;;;::::::::::::;
"o s
2 l.1.l!l.
!l ! l
[
"U.
l'.
w e
u e
O
_. l.!.
E o
l Z
l a
o w
o L.
'8 1
GA e
m e.*
2-*5 o
J
?~To 6
3;tt e
>=r e
e e
e e
..y 3
si c
e
=
I
- . g i
i g l 3
,J"3 ;
2, i
O I
e w
e g
," s I
e I
g e.-
8 o
[-
w I
a g
I 8
g I
i u
8 C
l l
2 s
I m
L _ _ i_
L._ _ J E-L/3 "E E'-
+2
's C
3#a"
-TCE Cu v
~~
m U
e u
w**
e r '="
t/')
2 "E
y
+..
1 I
V 1
I o
o i
I y
ed 6
1 I
g*
c
- y 8
l l
r=
3
- t..
l m
l O^
>a l e C W of I
s.
.o I
w=
t Oe E se o t_
b I
o-a to e
.e-U C3 "C o oc "e *a' Qg O 3 e.e uu
-~-
"e' E U pp l
r
=
Er:
? '.,0 i
=
C nn U C-
>c a
ww uf5 5
x 4
M Y
N.
e Y
I I 5 o
6 g
m I-m Q
em N
t
". 6 :::
-w e
L l
I i
ff6 5
~
~
5 s
a I
CONSERVATISMS i
IPWRS&BWRS E
O EFFECTIVENESS OF THE BACKUP SCRAM SYSTEM IS GIVEN RELATIVELY SMALL CREDIT.
ADDITIONAL CREDIT WOULD SHARPLY REDUCE THE CALCULATED FREQUENCY OF EXCEEDING THE GUIDELINES FOR THIS STUDY.
O IRANSIENT CHALLENGES TO THE SCRAM SYSTEM INITIATED FROM POWER LEVELS 25% ARE ALL TREATED AS WORST CASE, I.E., TREATED AS INITIATING AT 100% POWER.
5 0
AN ATWS IS ASSUMED TO LEAD TO UNACCEPTABLE PLANT CONDITIONS, IF THE ATWS OCCURS WITH THE MTC IN A REGIME WHICH DOES NOT HAVE DETERMINISTIC ANALYSIS TO SUPPORT IT.
O FOR PORTIONS OF PWR LIFE, THE MIC IS SUFFICIENTLY NEGATIVE TO WITHSTAND AN ATWS PLUS ADDITIONAL FAILURES IN BACKUP SAFETY SYSTEMS.
THE AVAILABLE DETERMINISTIC ANALYSIS DO NOT ALLOW THE INCLUSION OF THIS IN THE ANALYSIS.
E O
RPT MAY NOT BE NECESSARY FOR ARI TO BE EFFECTIVE AS l
ASSUMED IN THE ANALYSES.
O PEAK LOCAL SUPPRESSION POOL TEMPERATURE LIMITED TO 200* F.
b.
-