493rd Meeting - June 7, 2002
Official Transcript of Proceedings
NUCLEAR REGULATORY COMMISSION
Title: Advisory Committee on Reactor Safeguards
493rd Meeting
Docket Number: (not applicable)
Location: Rockville, Maryland
Date: Friday, June 7, 2002
Work Order No.: NRC-419 Pages 383-495
NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers
1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005
(202) 234-4433 UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
+ + + + +
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS (ACRS)
493rd MEETING
+ + + + +
FRIDAY, JUNE 7, 2002
+ + + + +
ROCKVILLE, MARYLAND
+ + + + +
The ACRS met at the Nuclear Regulatory
Commission, Two White Flint North, Room T2B3, 11545
Rockville Pike, at 8:30 a.m., Dr. George E.
Apostolakis, Chairman, presiding.
COMMITTEE MEMBERS PRESENT:
GEORGE E. APOSTOLAKIS Chairman
MARIO V. BONACA Vice Chairman
F. PETER FORD Member
GRAHAM M. LEITCH Member
DANA A. POWERS Member
VICTOR H. RANSOM Member
STEPHEN L. ROSEN Member
WILLIAM J. SHACK Member
JOHN D. SIEBER Member
ACRS STAFF PRESENT:
SAM DURAISWAMY Designated Federal
Official/Technical
Assistant
JOHN T. LARKINS Executive Director
SHER BAHADUR Associate Director
ROBERT S. ELLIOTT Senior Staff Engineer
MAGGALEAN W. WESTON Special Assistant
NRC STAFF PRESENT:
GOUTAM BAGCHI NRR
SUZANNE BLACK NRR
ED CONNELL NRR
JOHN HANNON NRR
GENE IMBRO NRR
PETER J. KANG NRR
GABRIEL KLEIN NRR
P.T. KUO NRR
KAMAL MANOLY NRR
JIM STRAISHA NRR
DAVID TERAO NRR
ERIC WEISS NRR
LEON WHITNEY NRR
HERMAN GROVES RES
J.S. HYSLY RES
AMARJIT SINGH RES
ALSO PRESENT:
FRED EMERSON NEI
PAUL GUNTER NIRS
H.D. THORNBURG CONSULTANT
ROGER HUSTON LSS
I N D E X
Opening Remarks by the ACRS Chairman . . . . . . 387
Proposed Rulemaking to Enforce National Fire
Protection Association (NFPA) 805,
"Performance-Based Standard for Fire
Protection for Light Water Reactor
Electric Generating Plants". . . . . . . . 388
Generic Resolution of Voids in the Concrete
Containment. . . . . . . . . . . . . . . . 478
Adjourn. . . . . . . . . . . . . . . . . . . . . 495
P R O C E E D I N G S
8:30 a.m.
CHAIRMAN APOSTOLAKIS: The meeting will
now come to order. This is the second day of the
493rd meeting of the Advisory Committee on Reactor
Safeguards. During today's meeting, the Committee
will consider the following: Proposed Rulemaking to
Endorse National Fire Protection Association Standard
805; Generic Resolution of Voids in the Concrete
Containment; Future ACRS Activities; Reports of the
Planning and Procedures Subcommittee; Reconciliation
of ACRS Comments and Recommendations; and Proposed
ACRS Reports.
This meeting is being conducted in
accordance with the provisions of the Federal Advisory
Committee Act. Mr. Sam Duraiswamy is the designated
federal official for the initial portion of the
meeting. We have received no written comments or
requests for time to make oral statements from members
of the public regarding today's session. A transcript
of a portion of the meeting is being kept and it is
requested that the speakers use one of the
microphones, identify themselves and speak with
sufficient clarity and volume so that they can be
readily heard.
Before we start, I'm very pleased to
announce that our own Dr. Powers was elected Fellow of
the American Nuclear Society recently.
(Applause.)
MEMBER KRESS: Well deserved.
CHAIRMAN APOSTOLAKIS: Well, deserved.
MEMBER POWERS: Thanks, Tom.
CHAIRMAN APOSTOLAKIS: The next topic or
the first topic is --
MEMBER ROSEN: Do you want to tell the
Members about the picture at 3 o'clock?
CHAIRMAN APOSTOLAKIS: Yes, at 3 o'clock
we will have our picture taken and at 1:30 there is
cake in the room next door celebrating somebody's
birthday who is 29 years old.
(Laughter.)
MEMBER WALLIS: Doesn't give us much
choice, does he?
CHAIRMAN APOSTOLAKIS: The first item of
the agenda is the Proposed Rulemaking to Endorse
National Fire Protection Association 805,
"Performance-Based Standard for Fire Protection for
Light Water Reactor
Electric Generating Plants." Mr. Steve Rosen is the
cognizant member.
Steve?
MEMBER ROSEN: Thank you, Mr. Chairman.
We had an exciting subcommittee meeting on June 4th
here in this room. It was well attended, scheduled to
last all day and contrary to the expectation
yesterday, that we did, in fact, stay all day and into
quite late hour last night on another subject, the
Fire Protection Subcommittee was able to wrap up its
work in half a day.
CHAIRMAN APOSTOLAKIS: Does this reflect
on the chairman of the subcommittee?
(Laughter.)
MEMBER ROSEN: I think it reflects
somewhat on the chairman's ability to run an effective
meeting, yes.
(Laughter.)
CHAIRMAN APOSTOLAKIS: So noted.
MEMBER ROSEN: I'm referring to the
subcommittee's chairman.
CHAIRMAN APOSTOLAKIS: I understand,
otherwise, we would not be having this discussion.
(Laughter.)
MEMBER ROSEN: We did, in fact, discuss
two topics at that meeting, not one as we have here
today. We talked about the Fire Protection
Association's 805 standard, but we also talked about
NEI 00-01, the Resolution of Circuit Failures document
that NEI is working on. That is not a topic of
today's meeting, but I thought I'd just mention that
that was discussed at the tail end of the
subcommittee. We'll come back to the subcommittee
later on.
The NFPA 805 is a standard that the
Committee has reviewed before and was critical of in
the basic sense that the previous versions did not
really move towards risk informing the fire protection
rules which, as you know, are very prescriptive. The
new standard is different and moves in a significant
way to react to comments that this Committee made in
1999 and I'll let the presenters tell you about that
and then I'll conclude.
So if we can go ahead with Mr. Weiss.
MR. HANNON: This is John Hannon in Plant
System Branch Chief. While Eric is setting up, let me
introduce Suzie Black is with us this morning. She's
the recently appointed Deputy Division Director for
DSSA. Of course, I'm with the Plant Systems Branch
and as mentioned, we're here this morning to brief you
on the status about risk-informed performance-based
rulemaking in the fire protection area.
We, as mentioned, had the opportunity to
brief the subcommittee earlier this week and look
forward to getting your comments and advice as we move
forward in the rulemaking process. We believe this
volunteer rulemaking sets the stage to improve the
coherency of our regulations in the fire protection
area and to the extent it is adopted by licensees, has
the potential to enhance our efficiency and
effectiveness.
With that, let me now turn it over to Eric
Weiss who will conduct the briefing for the staff.
MR. WEISS: Good morning. I'm Eric Weiss,
Chief of the Fire Protection Section. On slide 2, I
have a brief outline of the nature of today's
briefing.
(Slide change.)
MR. WEISS: On slide 3, let's begin.
(Slide change.)
MR. WEISS: As you know, Appendix R is
essentially deterministic regulation. Our existing
fire protection regulations have some rather
prescriptive requirements and the National Fire
Protection Association undertook the task of
developing a risk-informed performed-based fire
protection standard for the existing fleet of light
water reactors.
In February of 1999, the ACRS commented on
the development of that standard and was critical, in
part, because risk assessment as not allowed to alter
the basic requirements in many aspects. I want to
assure you that the staff has kept this in their mind
and we believe that we've been able to resolve that to
some extent which I'll describe in the nature of our
rulemaking which endorses the standard.
We issued a comprehensive reg guide on
fire protection, Reg. Guide 1.189 recently, and among
other things that Reg. Guide lays out the criteria for
an adequate fire protection program.
NFPA 05, as a risk-informed national
consensus standard was issued in February of 2001 and
it was developed in accordance with the approval of
the American National Standards Institute which means,
in part, that the makeup of the committee that
developed this standard had to meet the ANSI
requirements for balance on the committee.
Slide 4, please.
(Slide change.)
MR. WEISS: The rule to endorse NFPA 805
is consistent with the National Technology Transfer
and Advancement Act and OMB Circular A119 in the sense
that these laws and directives require federal
agencies to use national consensus standards in lieu
of agency developed or specific criteria when they
serve the needs of the agency.
NFPA 805 takes advantage of the advances
in PRA and fire science since Appendix R was issued
some 20 years ago. There have been substantial
advances in fire modeling and in PRA since then and
this is a rule that permits us to move into a risk-
informed performed-based area.
We can always accept exemptions to our
existing regulations at any time on whatever basis
licensees choose to justify those exemptions. They
can certainly be risk-informed performed-based, but
this is a rule that will permit licensees to move
forward in this area without exemptions.
Before proceeding further, I want to point
out that Appendix R and NFPA 805 achieve fire safety
through slightly different methods.
(Slide change.)
MR. WEISS: On slide 5 is a Ven diagram
which is not comprehensive in its depiction of the
differences, but more illustrative. I have a few
examples of this VEN diagram to show you that Appendix
R has a plant capable of going to cold shutdown within
72 hours following the event, but it does not apply to
all shutdown modes.
Conversely, NFPA 805 requires fire safety
in all operational modes and does not require the
plant to go to cold shutdown. It in effect requires
a hot shutdown because it requires a safe and stable
condition. There are other differences, but I just
wanted to make that point, that it achieves fire
safety in different ways.
MEMBER WALLIS: I guess the word shutdown
is missing after achieving in NFPA -- in the middle
there, "the plant from achieving shutdown" -- it has
to achieve something.
MR. WEISS: Yes, "Safe and stable
shutdown."
MEMBER WALLIS: Oh, "safe and stable
shutdown."
MR. WEISS: Yes, it says condition.
MEMBER WALLIS: Is missing from after the
word "achieving."
You can't achieve the fuel. You can't
achieve the fuel in a stable condition. I mean it's
got to achieve something. Achieving shutdown it must
be.
MR. WEISS: Slide 6, please.
(Slide change.)
VICE CHAIRMAN BONACA: I just want to ask
a question. Why now NFPA does not address the path to
cold shutdown?
MR. WEISS: Well, these standards are
designed to achieve fire safety in different ways and
I think everyone on the Committee knows that most PRAs
end at about 12 hours following the event. It would
be difficult to show a risk advantage. In addition,
I'm sure the Committee knows that there are more
systems available for maintaining hot shutdown.
MEMBER POWERS: I think one of the
motivations was particularly if you have a passive
plant like an AP1000, you really have troubles driving
everything to cold shutdown and in the event of an
off-normal event and I think people recognize that.
VICE CHAIRMAN BONACA: Okay.
MEMBER POWERS: And they were saying get
to a safe and stable condition.
VICE CHAIRMAN BONACA: Condition, rather
than --
MEMBER POWERS: With the AP1000, it may or
may not be cold shutdown and it certainly won't be
cold shutdown for some time period after the event.
VICE CHAIRMAN BONACA: Okay.
MR. WEISS: One of the chief advantages of
NFPA 805 is that it allows licensees to maintain
safety through more flexible, efficient and rational
processes. In other words, to use engineering as
opposed to meeting strict deterministic requirements.
We expect that this rulemaking will reduce
the number of exemptions and submittals, in part,
because the structure of the rule is such that
licensees can adopt the methods without making
submittals. It allows the use of risk insights, fire
modeling, science and engineering that's consistent
with NRC's outcome goals and it allows licensees to
focus their fire protection program on the most safety
significant issues.
MEMBER WALLIS: The previous method didn't
allow use of science and engineering?
(Laughter.)
MR. WEISS: The previous method required
that you met certain strict deterministic
requirements. For example, suppose a licensee goes
into a plant and discovers that a fire wrap that was
supposed to provide 1-hour barrier protection is no
longer capable of providing that one hour of
protection. Let's say for the sake of argument that
it's worth 40 minutes. Well, then the licensee is
confronted with a choice. They have to restore that
to the 1-hour condition or they have to apply for an
exemption. Under an 805 process, they could use a
fire model and engineering and say in this particular
area we can show that we only need 40 minutes. We
don't need an hour. And so engineering could
substitute for the replacement of the material and
that's the point I'm trying to make.
Obviously, engineering is involved in both
aspects, but there's much more flexibility associated
with being able to apply a fire model or use PRA as
opposed to meeting a strict deterministic requirement.
(Slide change.)
MR. WEISS: On slide 7, NFPA 805 allows
the transition of the existing Appendix R licensing
basis, including the exemptions and the General Letter
86-10 equivalencies. So for the most part, a
licensee's existing licensing basis would transfer
over. It allows future changes to the plant to be
either deterministic or risk-informed.
And at this point if the Committee will
permit me, I'm going to put up a diagram out of 805 on
the overhead projector. Let's see here if I can get
this right.
This is Figure 2.2 out of NFPA 805. At
the top, one enters the process and has certain basic
requirements one has to meet, but on the left hand
side there's a set of deterministic requirements, a
deterministic path and on the right there's a
performed-based path. Thank you.
There's the deterministic path and this is
the performance-based path.
The deterministic requirements as I'll
outline in a moment are very much like what's in
Appendix R right now. To go back to the hypothetical
example that I was using a minute ago where at some
point in the future a licensee discovers that fire
wrap is no longer good for an hour, he still has
available to him under the 805 process the option of
going to the deterministic path and restoring that
wrap to the 1-hour requirement. He isn't required to
do the analysis for the performed-based method.
That's an option. He can go either way.
MEMBER ROSEN: So to clarify that, first,
the licensee adopts 805 and does what he needs to in
the regulatory framework to do that. And then when he
finds a problem, he can use either a risk analysis or
the old deterministic basis. So there's two choices,
one big choice and then a potential for a whole lot of
little choices.
MR. WEISS: I think that's correct, yes.
CHAIRMAN APOSTOLAKIS: Well, actually,
there is no choice because right now they're supposed
to comply with Appendix R and that's a starting point.
MEMBER ROSEN: Right, but for a licensee
it ends up with a whole bunch of possible future
choices, not excluding using as Eric says the old
deterministic basis, if he wants to.
CHAIRMAN APOSTOLAKIS: That's right.
MEMBER ROSEN: So he gives up really
nothing.
CHAIRMAN APOSTOLAKIS: I'd like to
understand in this diagram, if one follows the
so-called deterministic approach, how can one to a
risk-informed change evaluation?
MR. WEISS: The deterministic approach is
not a risk-informed performed-based path.
CHAIRMAN APOSTOLAKIS: Yes, but you see
the way the boxes are, it implies I can one of two
things inside the big box, but then I can go on to
risk-informed change evaluation.
MR. WEISS: One can evaluate -- Ed, can
you help me out here?
MR. CONNELL: Sure. Of course, as you are
all familiar --
CHAIRMAN APOSTOLAKIS: Identify yourself,
please.
MR. CONNELL: This is Ed Connell from the
staff.
There are several deterministic options
like under 3G2 of Appendix R. You can put 1-hour
sprinklers, 1-hour 20 feet of separation or 3 hours.
All of those, while under Appendix R space were
considered equivalent, they are not necessarily
equivalent when assessing the risk. So if you made a
change under a deterministic approach, you still have
to assess the risk impact of that. We would expect it
wouldn't be significant, but consistent with the
risk-informed process, whenever you make a change, you
assess the risk, whether you're using a deterministic
approach or performed-based approach.
CHAIRMAN APOSTOLAKIS: So this diagram is
a bit misleading, is it?
MR. CONNELL: No, it's exactly accurate.
If you make a change under the deterministic approach,
you still do a risk change evaluation.
CHAIRMAN APOSTOLAKIS: Now where is that?
MR. CONNELL: It's at the bottom.
MEMBER SHACK: It's what you were just
pointing out, George. Both of them lead to that box.
MR. CONNELL: Right.
CHAIRMAN APOSTOLAKIS: No, only one leads
to that box.
MR. CONNELL: No, both.
CHAIRMAN APOSTOLAKIS: How can the
deterministic approach lead to that unless you put an
extra line there behind the --
MEMBER POWERS: George, once you have set
up your fire protection thing, you've done the
analyses, be they performed-based or deterministic,
then you have to do an overall risk assessment of your
fire protection system.
CHAIRMAN APOSTOLAKIS: Where does it say
that?
MR. CONNELL: The only risk assessment
that's required is that they -- when you make a change
to the plant, from whatever it is today, okay, you
make a change to the plant, whether you're using a
performed-based approach or a deterministic approach,
okay, you change it from one deterministic approach to
another deterministic approach, you still have to do
a risk change evaluation. You've got to look at the
change in risk as resulting from that change, if there
is any.
CHAIRMAN APOSTOLAKIS: My point is that
this diagram does not convey that. Of course that is
what you have to do.
MEMBER POWERS: But George, that's not the
issue. The issue is whether you -- whether the
standard is acceptable. Okay? I mean you have to
reach this diagram in the context of the standard
which is once you have done any kind of change, you
still have to do an overall risk assessment and see
how that changed the risk.
CHAIRMAN APOSTOLAKIS: We just said with
Steve that you have an option. But it seems to me
that if you decide to stay in the deterministic
approach, you cannot really justify a change.
MEMBER POWERS: Yes, you can. I mean --
I changed the way I have my stand pipes, okay?
CHAIRMAN APOSTOLAKIS: Right.
MEMBER POWERS: I did that with a
completely deterministic analysis, they have to be so
high and so big, things like that.
Now I do an overall risk assessment of my
fire protection scheme. How did that change change my
risk?
CHAIRMAN APOSTOLAKIS: And what I'm saying
is that this diagram doesn't say that, but anyway, if
everybody else understands it, I suppose I'm a
minority of one.
MEMBER POWERS: If you're relying on just
the diagram without understanding the way the standard
is written, okay, then that doesn't communicate to
you. But within the standard, it seems to me it makes
perfect sense.
CHAIRMAN APOSTOLAKIS: So even at the top
you have already done a risk assessment.
MS. BLACK: George, can I point out that
there's an arrow going into the large box and the
large box contains both the paths.
CHAIRMAN APOSTOLAKIS: Right.
MS. BLACK: And then there's another arrow
coming out of that large box.
CHAIRMAN APOSTOLAKIS: Right.
MS. BLACK: If you look at it that way,
then either side -- I agree, it's not the best drawing
in the world to depict it, but there's an arrow there
that comes out of the big box that includes both
paths. You get into the big box at the top without an
arrow going into anything in the big box either.
MEMBER WALLIS: Or coming out.
MS. BLACK: Or coming out. See, that's
the flaw in the drawing.
CHAIRMAN APOSTOLAKIS: Inside the box, I
get the sense that I can go either way, but then when
I exit the box, I realize that I really have to do
that -- but that's okay. I mean if everybody thinks
that's obvious.
MEMBER WALLIS: George, it's like
university administration. It's in the big box. It
really doesn't have arrows going in or coming out.
CHAIRMAN APOSTOLAKIS: So 1.174 will be
somewhere there in the evaluation of the change? Is
the change acceptable? Is that where 1.174 would come
in?
MS. BLACK: No. This is Suzanne Black.
1.174 gives you the same criteria, basically, they're
used to acceptability, but that's -- but that one,
1.174 is for license amendments.
CHAIRMAN APOSTOLAKIS: Is what?
MS. BLACK: For license amendments, but
the same basic concepts are used.
CHAIRMAN APOSTOLAKIS: If I don't use
1.174, how can I decide whether the change is
acceptable?
MR. CONNELL: Well, within NFPA 805, okay,
it says it uses CDF and LERF to measure the risk
impact.
CHAIRMAN APOSTOLAKIS: Right.
MR. CONNELL: So you look at a delta CDF
and a delta LERF.
CHAIRMAN APOSTOLAKIS: Right.
MR. CONNELL: And consistent with 1.174,
the increase in risk should be small; defense-in-depth
has to be maintained; and safety margins have to be
maintained.
CHAIRMAN APOSTOLAKIS: Right. So that's
where it is?
MR. CONNELL: That's all in that little --
that's in that little box -- where it says
risk-informed change evaluation, all that is explained
in the text of the standard.
CHAIRMAN APOSTOLAKIS: So 1.174 is there?
MR. CONNELL: Well, 1.174 is referenced in
the standard, but 1.174, if you look at the scope and
the application of 1.174, it's only for license
amendments that are submitted by licensees to the
staff for review and approval.
Under 805, these changes would be made
without NRC prior review and approval.
CHAIRMAN APOSTOLAKIS: Other criteria that
would be used --
MR. CONNELL: The criteria is the same.
CHAIRMAN APOSTOLAKIS: Yes, so 1.174.
MEMBER SIEBER: 50-59 for fire.
VICE CHAIRMAN BONACA: Although 50-59 does
not use the criteria. It says negligible.
MEMBER ROSEN: But it's like 50-59 in the
sense you can go ahead without prior NRC approval.
CHAIRMAN APOSTOLAKIS: Right, right.
MEMBER LEITCH: Once a licensee selects
one method or the other, does that determine the
approach he must always take or can this be decided on
a case by case basis whether to use the deterministic
approach or the performed-based --
MR. CONNELL: This can be done on a fire
area by fire area basis.
MEMBER LEITCH: Fire area by fire area.
CHAIRMAN APOSTOLAKIS: In some areas you
are risk-informed and some you aren't.
MEMBER RANSOM: How do you ever get out of
this process? It seems like all paths lead back to
the big box.
(Laughter.)
MEMBER SHACK: That's job security.
MR. CONNELL: There's a feedback process
here which we don't have right now and if you make a
change, let's say you use a specific assumption in
your performed-based approach, you assume that the
maximum combustible loading in this area is X value.
Well, a couple of years from now you increase the
combustible loading in that area to three or four
times that X value. Under 805, that change would have
to be factored back into the big box, so you see that
now that I've made a change to my plant, how does that
impact my previous analysis?
MEMBER ROSEN: We can go on, I think.
MR. CONNELL: Okay, one of the essential
elements of a risk-informed performed-based method is
that it can have a change control process and as we've
just been discussing, NFPA 805 has such a change
control process.
Now, in response to a key concern of our
stakeholders and of this Committee, we incorporated
into the rule that risk-informed performed-based
methods may be approved by the NRC.
MEMBER POWERS: If NRC was looking at a
proposed performed-based change, would they ask for a
performance indicator? Quite frankly, I don't see how
they couldn't.
MR. CONNELL: Well, there was a separate
effort, I think you're aware of developing performance
indicators and the conclusion related to fire
protection was there really isn't any good performance
indicators for fire protection. So the feedback
approach in 805 really addresses the assumptions that
are used in a performed-based analysis. I assume my
sprinkler system is reliable. I assume this is the
combustibles I'm going to have. I assume this is the
heat release rate of the geometry of the material. So
that's the feedback process we have in 805 which is
different than performance indicator process. The
only performance indicator we really have is like fire
frequency and that's not very reliable for assessing
performance of the fire protection program.
MEMBER POWERS: It's not especially
useful. I mean that's where I'm struggling a little
bit with performed-based. How can anything be
performed-based? Maybe it is as you say that indeed
I assume my sprinkler has this reliability and over
the course of time I find the number. I get an
indication of the number of times that I've had to fix
it. I know how often it's down because I'm
maintaining it just in the course of normal events,
but sometimes I have to fix it.
It seems to me that you get a -- the
number you get out of that is not wildly reliable.
Unless you've got some process by which I find I have
to fix my sprinkler system. I suddenly have
discovered a day that I have to fix my sprinkler
system and so now how many weeks has it been
inoperable and I didn't know about it?
MR. CONNELL: Well, of course, all the
numbers wouldn't be very reliable. That's why we have
defense-in-depth. That's why there's a fundamental
aspect of 805. So we don't place reliable on safety
solely on that sprinkler system. We have other things
that are there. So the sprinkler system doesn't
perform as expected we still have some level of
safety.
MEMBER POWERS: I will grant we have that.
I'm still struggling with what's performance, how I
consider something performed-based in this.
MR. CONNELL: Well, the performed-based
approach to 805 is you can use a performed-based
approach to meeting the specific criteria. In the
past, we said okay, you're fire safe if you have a
1-hour barrier and sprinklers. Now we can say all
right and you have that train, it's used for DK heat
removal, let's say. Well, instead of having 1-hour
and sprinklers for DK heat removal, we can say hey,
I've assessed this fire area. This is the act. I've
modeled it. I've assessed the frequency of fires in
this area. I've assessed the damage threshold, the
fragility of the components I'm looking at and this,
I don't need one hour. Maybe I need 20 minutes.
Maybe I don't need anything. Okay? This is the
approach that's an 805, to demonstrate that that
system is still going to be maintained free of fire
damage.
MEMBER POWERS: See, what I think you're
getting to is in reality, I echo Dr. Kress here, that
there is no performed-based system that you either
have a probabilistic system or a deterministic system
here. And what you've outlined is an analysis that is
essentially probabilistic and if I were Dr. Kress I
would sit there and say okay, what is your acceptance
criteria and for this probabilistic analysis. In
other words, the guy says gee, I don't need an hour.
I only need 27.5 minutes. And Dr. Kress would say to
what level of confidence do you need 27 minutes?
MR. CONNELL: Well, we address that as
well in the standard. If you're using a fire model
and you're using a specific fire scenario, you say
well, this is my expected fire scenario and let's say
it was 10 gallons of Heptane, okay? I've looked at my
area and this is what I typically have in there.
You also under 805 have to look at what
causes damage. So if 10 gallons of Heptane is your
expected, that doesn't cause you to exceed, not meet
the performance criteria and you find out well, where
does it cross that threshold where I no longer meet
the performance criteria. Let's say that was a 100
gallons of Heptane. Then you know what your margin
is. You say okay, this is what I expect is 10. I
don't get damage until I have 100, so I have
confidence in that margin of safety and that addresses
the uncertainty.
MEMBER POWERS: See, I think it's where
Dr. Kress would be really confused. Because you've
come along and you've said okay, I've done this
probabilistic thing with 10 gallons per minute and I
can conclude that it does cause no damage. No you
didn't. You concluded as a point estimate that you
don't get damage. This is in reality had you gone
through and done the analysis correctly with lots of
attention to certainty and phenomena you have
concluded to a 92 percent confidence level, I don't
get damage with 10 gallons. And then you would say is
that good enough?
MEMBER KRESS: And why is that good
enough?
MR. CONNELL: I guess as a surrogate for
that we find out where we do get damage and if we
have, like I said, if the margin is from 10 gallons to
100 gallons, I think everybody would agree that we
have adequate safety margin.
MR. HANNON: Dr. Powers, this is John
Hannon. I would just like to point out that the Fire
Risk Research Program does have the development of
performance indicators as part of that activity.
MEMBER POWERS: They can research until
the cows come home. Everybody has looked at this
thing and they come back and say there just aren't any
good ones and great confidence in the research guys
and they may find them, but the fact is we don't have
them right now and the fact is I think Dr. Kress is
correct. There is no performance in here. You either
have got a probabilistic side or a deterministic side
and when you go down the probabilistic side, we've got
this problem, what's acceptable? And doing a point
calculation is just never acceptable. Never
acceptable.
MEMBER ROSEN: Well, I think I'd like to
try to steer us back into the slow lane and go on in
this general direction.
CHAIRMAN APOSTOLAKIS: The previous slide,
the last bullet said something I didn't understand.
Can we go back to 7?
What does that mean, new risk-informed --
MR. CONNELL: Risk-informed performed-
based methods. Yes, I'm going to address that.
CHAIRMAN APOSTOLAKIS: "To be approved",
is that what it says?
MR. WEISS: Yes, to be approved. I'll
outline a little more in that, but let me touch
briefly on it now since you bring it up. Our
stakeholders have made clear that the requirements in
Chapter 3 contain a large body of deterministic
requirements. It's on one of my slides that follow
and that's in a large part because there aren't risk-
informed performed-based methods for these things
which I'm going to cover in a future slide. But the
concern was, and I think part of the concern of the
Committee was that if this is going to be a
risk-informed performed-based method, should it not
allow the use of future risk-informed performed-based
methods. So we tried to build into this rulemaking a
provision that we could accommodate those methods when
the staff approved them.
I think it will become clearer later on.
If it's not, then please ask again. But could I go on
to slide 8, please?
(Slide change.)
MR. WEISS: As I indicated before, 805
allows either a deterministic approach or the
risk-informed performed-based approach. There are two
paths and the deterministic requirements read very
much like what is in Appendix R right now.
CHAIRMAN APOSTOLAKIS: So I can still come
to you using Appendix R methods and request an
exemption, correct?
MR. WEISS: Under any rulemaking you can
always request an exemption, but under this
rulemaking, if 805 became part of a plant's licensing
basis, the point I'm trying to make is a licensee
could say okay, I'm going to put in a 3-hour barrier.
CHAIRMAN APOSTOLAKIS: Right.
MR. WEISS: As opposed to saying I'm going
to do a fire model and a PRA to show what the barrier
should be.
CHAIRMAN APOSTOLAKIS: right, but I mean
at some point you require 20 feet separation and your
licensee can come and argue that in this particular
case 18 feet is good enough and you do your
engineering evaluation so you say it's good enough,
right?
MR. WEISS: Yes, and the huge advantage is
that it's no longer an exception. It's being done
under the 805 process.
CHAIRMAN APOSTOLAKIS: Oh no, I don't want
to do 805. I just want to come to you and do it the
old way. I can still do that?
MR. WEISS: Yes. A licensee can submit an
exemption right now under Appendix R under any basis,
under a risk-informed basis, under a performed-based
and the staff will review those individual plan
exemptions.
CHAIRMAN APOSTOLAKIS: Now, is there some
analysis somewhere that can tell us -- you have
approved numerous exemptions within Appendix R.
MR. WEISS: Not 900.
CHAIRMAN APOSTOLAKIS: 900, for the last
20 years or so. Do you have any idea what the implied
delta CDF and delta LERF was?
MR. WEISS: I can say that the issue of
the cumulative effect of these exemptions did come up.
Ed, can you help me out here?
MR. CONNELL: Yes, in conjunction with the
Office of Research and a contract with Sandia, we
looked at the ten highest reported fire induced CDS
resulting from the IPEEE program. And then we looked
at all the exemptions that were granted to those 10
plants. And the conclusion was that the exemptions
granted had little or no risk significance.
CHAIRMAN APOSTOLAKIS: So that implies
then that if I follow the performed-based approach,
presumably it can ask for me.
MR. CONNELL: More what? You don't have
to ask for anything under the performed-based
approach. I can.
CHAIRMAN APOSTOLAKIS: You say that the
largest approved Appendix R-related exemption led to
negligible delta CDF, right?
MR. CONNELL: Right.
CHAIRMAN APOSTOLAKIS: In the future, if
I wanted to do something more serious, then I can go
to them for CDF.
MR. CONNELL: I guess it depends where you
are right now.
CHAIRMAN APOSTOLAKIS: That's right.
MR. WEISS: Slide 9, please.
(Slide change.)
MR. WEISS: Here's a list of the
requirements in Chapter 3, the fundamental fire
protection elements and right now these are somewhat
prescriptive, but as I indicated before we've built a
provision in the rule to allow for the staff to adopt
new risk-informed performed-based methods should they
be justified.
Slide 10?
(Slide change.)
MR. WEISS: The way the rule is structured
is that it will be an amendment to 10 CFR 50.48 and
adoption of the provision that puts you into 805 is
voluntary. This was very important to our
stakeholders and that's the way the rule is
structured. Licensees can choose to stay under their
existing Appendix R licensing basis and the question
came up in the subcommittee why would one switch over?
How many people would switch over? How many licensees
would switch over and I'm going to let NEI address
that a little more definitively, but I'll say
conceptually I can understand why a licensee that has
a perfectly good licensing basis and no reason for
change might very well, as a matter of fact, I might
expect most licensees to stick with Appendix R until
such time as they see an advantage to solving a
problem.
CHAIRMAN APOSTOLAKIS: I think that's what
I just said. They feel they can get an exemption or
a change approved by the NRC easier by doing the risk
argument, giving a risk argument, especially if you
say that the old exemptions led to negligible delta
CDF, if I want my delta CDF not to be negligible now,
I have a better way of doing it because my chances
that you would approve it under the old Appendix R are
very, very small. That's all they have. This is
really the appealing feature of this which leads me to
another thing. Is the industry still or NEI still
saying that the industry will not use this? I mean --
MEMBER ROSEN: I think we have Fred
Emerson here who is going to address that comment,
question. It came up very hard in the subcommittee
and he's got a number of points to make on that
subject.
CHAIRMAN APOSTOLAKIS: Okay.
MR. WEISS: As I tried to outline before,
the existing licensing basis configuration and
procedures largely convey to the new risk-informed
performed-based environment. The way the rule would
work is that licensees would document and retain the
records on site for inspection. The reactor oversight
process would monitor future changes and NRC would be
allowed to approve new risk-informed performed-based
methods in the future on licensees, may use the
appendices of 805 which are there for information.
Slide 11, please.
(Slide change.)
MR. WEISS: This is one of NRC's first
risk-informed performed-based rules, not the first,
but one of the first and NEI endorsed this rulemaking
in September of 2001. We recognize the key to
successful implementation of this approach is the
development of appropriate regulatory guidance. NEI
has agreed to develop a guidance document that we
could then endorse in a Reg. Guide. I might also
point out that NFPA 805 addresses the existing fleet
of light water reactors. There is a separate standard
804 for the advanced LWRs, but the staff has written
the NFPA and asked them to develop a new NFPA standard
to address advanced light water reactors and gas
reactors, other advanced reactors in a risk-informed
performed-based manner.
Slide 12?
(Slide change.)
MEMBER ROSEN: Why did you limit that to
just gas and light water reactors?
It says "Future NFPA standard to address"
--
MR. CONNELL: I know --
MEMBER ROSEN: "Advanced light water
reactors and gas reactors."
MR. CONNELL: It's not limited to the
light water and gas reactors.
MR. WEISS: It's all advanced reactors.
MR. CONNELL: It's all advanced reactor
designs.
MEMBER ROSEN: We had a discussion
yesterday that there are advanced reactor designs in
a Generation 4 program that used neither water nor
gas.
MR. CONNELL: Right, and the standard --
right now, the standard is just an idea, but the
intent was to address all advanced reactor designs.
MEMBER ROSEN: Okay.
CHAIRMAN APOSTOLAKIS: So I can see how
the decision whether the proposed changes are
acceptable depends on light water reactors. This is
Regulatory Guide 1174 gives delta CDF and delta LERF.
Are there any other parts of the standard that would
be different for advanced reactors besides the
acceptance criteria?
MR. CONNELL: Well, the performance
criteria outlined in several Commission SECY papers
for advanced reactors is different than what we have
for the existing fleet for fire protection.
CHAIRMAN APOSTOLAKIS: Can you give me an
example?
MR. CONNELL: Oh, you have to assume all
equipment in a fire area is rendered inoperable and
re-entry for operation repair is not permitted, so the
equivalent of 3G2 of Appendix R would not be allowed
for any of the advanced reactors. It's not allowed
for the CE system 80. It's not allowed for the GE
ABWR. It's not allowed for the AP 600 and it won't be
allowed for the AP 1000.
CHAIRMAN APOSTOLAKIS: That's Appendix R
type of thing?
MR. CONNELL: No. No. Appendix R, 3G2 is
for redundant systems located in the same fire area.
You're not allowed to have redundant systems located
in the same fire area with the advanced reactors.
That's why 805 wouldn't be applicable to
the advanced reactor designs.
Most of the issues we deal with are
related to redundant systems located in the same fire
area.
CHAIRMAN APOSTOLAKIS: But would the
advanced reactors have some specific features that
would require different approach? It seems to me the
basic approach that you have here is --
MR. CONNELL: Well, there's a different
approach like the AP600/1000 safe shutdown for them,
unlike the current fleet is not cold shutdown. They
don't get the cold shutdown.
CHAIRMAN APOSTOLAKIS: That's what it
says. Isn't that what Professor Wallis asked? That's
what this says that maintain the fuel in a stable
condition.
MR. CONNELL: Safe and stable condition,
right.
Yes. 805 -- I was talking about for
Appendix R plan, the current fleet --
CHAIRMAN APOSTOLAKIS: No, I'm talking
about 805.
MR. CONNELL: All right, 805, yeah, that
would be consistent. But most of the issues we're
relating to in the current fleet relate to redundant
systems located in the same fire area. Right, an
that's what most 805 addresses, but the administrative
controls, the fire brigade, all that kind of stuff, of
course, would be very similar.
MEMBER ROSEN: You have about five more
minutes, Eric.
MR. WEISS: Okay, slide 12 is the
schedule. We're before you today, we're scheduled to
go to CRGR on the 11th. The proposed rule is due in
front of the Commission in July. The proposed rule
will then be published in the Federal Register for
public comment for a period of one month. I might say
that we've been making various drafts of this
regulation available on the web and we informed the
Commission of that.
The final would go to the Commission 15
months after the close of the public comment period on
the proposed rule and the final rule would be
published in the Federal Register one month after the
staff requirements memorandum.
MEMBER ROSEN: What makes it 15 months
mandatory? It seems like an awful long time after the
close of public comments before you present it to the
Commission.
MR. WEISS: I can tell you there are some
rules that I've been associated with that went on for
seven years and never did see the light of day. My
first job in the Commission in 1976 was to lay out the
procedures for rulemaking and I used to be in charge
of laying out the procedure in the Green Book and what
was then the Office of Standards Development. In
those days, a rough rule of thumb was that it took
about a year to get a proposed rule out and about a
year to get a final rule out, but that was a rule of
thumb. Controversial rules easily go more than a
year. This is -- this schedule that we developed for
the 15-month schedule was not developed by mere
judgment as my previous comments might apply. We've
actually got a very detailed schedule laid out with
what we believe are realistic, not conservative,
realistic elements of each of the steps involved. So
I guess there is no simple answer to your question
other than for me to show you the schedule, but I can
tell you from a feeling point of view I've been
involved in four rulemakings in my 25-year career with
the NRC and some have gone on forever and two of them
never saw the light of day, two of them died after
years of controversy.
MEMBER ROSEN: We don't want this to die
and neither does the Commission and neither does the
stakeholders, neither does the ACRS. But I would be
interested in having you show me this 15 months packed
with activities after most of the music has been
played.
MR. WEISS: We can certainly do that.
CHAIRMAN APOSTOLAKIS: Well, after the
music, they go to a restaurant --
MEMBER KRESS: Have a cigarette, coffee.
CHAIRMAN APOSTOLAKIS: A nice cigar.
MEMBER POWERS: Fifteen months is what I
would call leaning forward in the trenches schedule
quite frankly.
MR. WEISS: Winning what?
MEMBER POWERS: I mean it sounds like
they're fairly optimistic. There is a lot that has to
be done. I'd like to see the steps.
MR. WEISS: We will do that. Slide 13.
(Slide change.)
MR. WEISS: We believe that this
rulemaking is an important part of a regulatory
framework that will move fire protection forward into
the
risk-informed performed-based arena. It certainly is
not the whole answer. There has to be regulatory
guidance. There has to be inspector guidance and
inspector training.
CHAIRMAN APOSTOLAKIS: Regarding guidance,
is there any guidance now as to what risk methods are
acceptable? If I came to you and I'm saying I'm
implementing 805 and I calculated delta CDF this way.
Are you going to check that and see whether what I did
was right or are there methods that are acceptable or
methods that are not acceptable?
MR. WEISS: I think maybe Fred could tell
you what they have in mind in the early stages of the
guidance document, but one of the things that they had
broached with us was endorsing an ANS standard that's
to be developed.
CHAIRMAN APOSTOLAKIS: For external
events? An ANS standard for external events.
MEMBER ROSEN: I think you want them for
fire and PRA too.
CHAIRMAN APOSTOLAKIS: ANS has developed
standards for earthquakes, fires, tornadoes. That's
what I mean by external events.
MR. WEISS: So we see this rulemaking as
a necessary first step in providing an opportunity for
licensees and NRC to be more efficient and effective
in this regulatory environment. That's the conclusion
of my presentation.
MEMBER LEITCH: Back on Slide 9 you have
a list of fundamental fire protection elements that
are in Chapter 3 of 805, are those deterministic
things? Could you say a little more about that?
MR. WEISS: I think Ed could give you the
details, but pick an example out of the air. Fire
brigade is five people.
MEMBER LEITCH: Right.
MR. WEISS: We don't have a method of
calculating that the fire brigade should be 4.2 people
right now, but conceptually, in the future, if there
were such a method and the staff approved it, that we
could incorporate that in this rulemaking. We have a
provision in the rulemaking that allows the Agency to
adopt it.
MEMBER LEITCH: But at the moment it still
says five people.
CHAIRMAN APOSTOLAKIS: But I can come to
you and show that I have other compensatory measures
that would justify having only four. I mean that's
the whole idea of 805, isn't it?
MR. WEISS: Yes.
CHAIRMAN APOSTOLAKIS: Instead of the
five.
MR. WEISS: Right now, 805 says five and
we haven't approved a method of calculating. You can
always apply for an exemption and Ed can speak to this
in more detail, but I think we've accepted exemptions
on a wide range of deterministic requirements, but the
idea behind the rule is to get out from underneath the
exemption process because that's relatively
inefficient and to have risk-informed performance-
based methods that licensees can adopt without making
even submittals to the NRC, simply put it in the file
draw and then we come around and make sure that they
use the right method, that the people who used it were
qualified and so forth and so on and that will make
for a much more efficient process and I think a more
rational process.
MEMBER LEITCH: But in the case, for
example, numbers of members in the fire brigade, a
licensee couldn't just on his own based on
risk-informed performed-based, decide that he only
needed four, put that documentation in the file and --
MR. WEISS: Not without making an
exemption request right now. But conceptually in the
future if there were such a way, then the staff could
adopt it and then that process could go forward
without an exemption.
MEMBER LEITCH: But that would be a
subsequent change to 805?
MR. WEISS: Well, we have a provision in
805 that allows us to adopt these new methods. That
was one of the stakeholder's key points and we thought
we were being responsive to the Committee, to the ACRS
in that regard as well.
CHAIRMAN APOSTOLAKIS: So right -- it's
kind of new to me. Right now, I can -- these bullets
that you have on Slide 9, these are requirements. It
says the fire brigade has to have five people, for
example.
MR. WEISS: Yes.
CHAIRMAN APOSTOLAKIS: And I cannot use
805 as it is now to do a calculation to show that I
can live with four because I don't have a method that
you have approved.
MR. WEISS: That's right.
MEMBER ROSEN: That's right. But if
someone came up with a method and it was peer reviewed
and discussed in the industry and the NRC looked at it
and adopted it, said yes, that's correct, it's a good
method, then the licensees who had adopted 805 could
use it.
VICE CHAIRMAN BONACA: A better example
would be if you have a number of fire extinguishers,
fire alarm protection, can you use 805 to assess, for
example, more risk significant areas where you would
concentrate more detection devices or suppression
devices?
MR. CONNELL: Yes, this is Ed Collins,
Staff again. 805, as far as the things that are
related to fire protection systems and features, when
your performed-based or deterministic approach says
you need a specific fire protection feature, a
detection system, a suppression system, then Chapter
3 of 805 says okay, if you're going to put this thing
in, it has to be designed, installed and maintained in
accordance with the applicable NFPA standard for that
system. So that's where it gets quasi-deterministic.
In other words, you say, okay, I need this system.
Under Appendix R it said you had to have this system
and you had to design, install and maintain it in
accordance with the applicable NFPA code. Under 805,
you decide whether you need it or not. If you do need
it, then you have to design and install and maintain
it in accordance with the applicable --
VICE CHAIRMAN BONACA: You don't have a
graded, more aggressive --
MR. CONNELL: There is no performed-based
risk-informed way of designing a sprinkler system.
There is no risk-informed performed-based way of
designing a detection system.
VICE CHAIRMAN BONACA: There is a
risk-informed way of determining which areas --
MR. CONNELL: Whether you need it or not,
right. That's correct.
VICE CHAIRMAN BONACA: Which area is more
important than others?
MR. CONNELL: Right, whether you need a
system or not is what 805 allows you to determine in
a performed-based risk-informed approached, whereas
Appendix R didn't allow you to do that.
CHAIRMAN APOSTOLAKIS: This thing about
the approved methods, maybe I don't understand it very
well. I mean in a risk assessment, there are all
sorts of models and assumptions that one has to make,
you know, in a fire risk assessment or any risk
assessment. And this notion of pre-approved methods
is not clear to me. In other words, I model the
susceptibility to damage of the insulation of the
cable in a certain way. I don't think that's a
standard way of doing it. Some people might say you
know, here's a temperature that applies. Somebody did
some experiments, so I'm going to use it as the limit
and the probability of exceeding it is such and such.
Somebody else might do detailed thermal calculations,
you know inside the insulation and go more into the
physics. And this is just an example.
Now what does it mean that there have to
be pre-approved methods? I mean that doesn't --
MEMBER ROSEN: Let me try on this one and
see if I have it right because that's a good question.
The answer is that that's what the implementation
guidance will say. It will say what the methods are
and what's allowed and what's not. And the staff will
approve that guidance in a regulatory guide. Is that
the right answer for this question?
CHAIRMAN APOSTOLAKIS: Well, I don't know
to what extent you can do that. That's my point.
MR. CONNELL: Well, the appendices in 805
don't tell you how to do these things. What they do
is they say here are the things you need to address in
order to do them. 805 allows you flexibility in
whatever the particular method that you use, okay, so
there's no method, a step-by-step cookbook, whatever
you want to call it approach in 805, similar to like
the five methodologies basically a cook book approach,
okay?
Let's say five methodologies would be
enhanced beyond because it was only intended to look
at severe accident vulnerabilities. Well, let's say
we had an enhanced five methodology that could be used
for regulatory compliance. That could be approved by
the staff. That would be in the NEI guide.
CHAIRMAN APOSTOLAKIS: Is it correct then
to understand that if I do something that I think is
innovative, the first time I do it, I have to come to
you?
MR. CONNELL: That's correct.
CHAIRMAN APOSTOLAKIS: But after that I
can use it, if you approve it.
MR. CONNELL: That's correct. That's the
intent.
CHAIRMAN APOSTOLAKIS: That makes much
more sense.
MEMBER ROSEN: That's the idea.
CHAIRMAN APOSTOLAKIS: So the first time
I do something, you guys should be aware of it and say
we like it, we don't like it, change it this way.
MR. CONNELL: That's right.
MEMBER ROSEN: And the reason for that, of
course, was to allow the state of the art to progress,
something we've --
CHAIRMAN APOSTOLAKIS: It puts bounds on
our --
MR. CONNELL: And the standard was written
with that in mind. That's why we didn't prescribe a
method because when every six months or whatever we'd
have to revise the standard.
VICE CHAIRMAN BONACA: Just to understand
the limit. I was pursuing that before. I didn't
understand it, so I assume that right now, Appendix R
says this area has this single safety category 1
system, therefore you have to have certain protection,
fire protection because it's important. My PRA says
that system is not risk significant.
And if I make changes which is the grade
the fire protection in that particular area, based on
the PRA?
MR. CONNELL: You'd be able under the
umbrella of NFPA 805, you would be able to relax the
fire protection systems and features provided,
provided you meet all performance, goals, objectives
and criteria; provided you still maintain
defense-in-depth, provided adequate safety margins are
maintained. All that good stuff, you would be allowed
to relax the fire protection.
Today, you can do that with prior staff
review and approval through the exemption process
outlined in 50.12.
VICE CHAIRMAN BONACA: Yes, thank you.
MEMBER ROSEN: Okay, Eric, can you wrap it
up?
MR. WEISS: I'm essentially done.
MEMBER ROSEN: Okay.
MR. WEISS: I'll turn it back to you.
MEMBER ROSEN: Thank you very much for a
good presentation.
Now we will hear the industry's
perspective on this proposed revision.
Fred Emerson from NEI.
CHAIRMAN APOSTOLAKIS: Now the
performance, when we say risk-informed performance,
the performance is at delta CDF, I suppose? Is that
what the performance is?
MEMBER WALLIS: Could you move your
microphone?
MEMBER SHACK: You end up then computing
a delta CDF when you're done.
CHAIRMAN APOSTOLAKIS: And this is my
performance measure, because I make my decision using
those.
MEMBER SHACK: You make your decision.
CHAIRMAN APOSTOLAKIS: Okay, sorry, Fred.
MR. EMERSON: Good morning. I'm Fred
Emerson from NEI. I'm the Fire Protection Project
Manager for fire protection issues. And I'm happy to
have the opportunity to come speak to the full
Committee. The presentation that I have is slightly
modified from the one that I gave on Tuesday to the
Fire Protection Subcommittee.
Next slide, please.
(Slide change.)
MR. EMERSON: The topics that I'm going to
cover you see on the slide. I'd like to provide a
little bit of background to provide some context. I'd
like to give, provide the fundamental industry
positions that led to our support for the rulemaking,
the current -- minute or two on the current rule
language, indicate what we're going to cover in the
implementing guidance and close by trying to address
a topic that the subcommittee addressed or wanted some
additional amplification on on how do we move forward
and who's going to use the standard or the rule when
it becomes a rule.
Next slide, please.
(Slide change.)
MR. EMERSON: Eric described the fact that
this was developed by an NFPA Committee. If was their
Technical Committee on Nuclear Facilities and it
comprised a several year effort. The industry and the
staff were both heavily involved in this activity and
a lot of good effort was put forward on both sides to
try to make this a useful standard.
When the final product was approved by the
industry in the fall of 2000, the industry still had
some concerns over what was in the final rule, the
final standard and as I understand, the NRC did as
well. And these concerns were to be dealt with in the
rulemaking process.
Next slide, please.
(Slide change.)
MR. EMERSON: When the rulemaking became
a reality, industry agreed to support the rulemaking.
We had several concerns. I'd like to just spend a
second on those.
The first one was the use of performed-
based methods to address Chapter 3. We just spent
some time talking about that. We felt that there
should be an allowance for the use of performed-based
methods to address these very deterministic
requirements of Chapter 3. Even if there were no
specific elements available, we felt that to support
a performed-based risk-informed standard there needed
to be an allowance for the use of those methods
throughout the use -- throughout the standard as a
whole, whereas the standard itself and Chapter 3 now
specifically prohibits that, so we sought an exception
to allow those.
We wanted to allow the use of docketed
licensing bases as previously -- instead of previously
approved alternatives which is also currently the
language of Chapter 3 because we felt that there were,
the concept of previously approved was fairly vague.
When you have an SER covering a specific topic, the
SER may be very general or very specific and if it's
very general, you often don't, cannot pinpoint whether
something was previously approved or not, so we had
submitted some alternate language for staff
consideration.
The third concern was that we -- that the
NRC perform a review of performed-based methods
instead of the NFPA Technical Committee and the staff
agreed to address that concern.
Another issue that I didn't mention on the
slide and that the subcommittee asked me to address
was the use of --
CHAIRMAN APOSTOLAKIS: I'm sorry, Fred,
why did you have to request that last one? Isn't that
what the NRC is supposed to do?
MR. EMERSON: Initially, there was a
discussion on the staff's part of allowing the NFPA
Technical Committee to review proposals and we thought
that was incorrect.
CHAIRMAN APOSTOLAKIS: Of course.
MR. EMERSON: This is really a dead issue.
This has been discussed and resolved. I'm just
pointing out what our initial --
CHAIRMAN APOSTOLAKIS: So the resolution
is that the NRC will review?
MR. EMERSON: Yes, that's correct.
Another issue that's not on the slide is how NEI 00-01
was to be used and Eric touched on that earlier. That
document is intended to be a risk-informed method for
resolving the circuit failures issue which we've
discussed before in front of this Committee. This is
going on in parallel with the NFPA 805 rulemaking. It
will be, we believe, will be resolved well before the
rulemaking is final and we would like to see that
method allowed as a one method of performing
engineering analysis for circuit failures as one
element of a licensee's risk-informed program.
We understand that that will be allowed
within the regulatory framework for 805.
Next slide, please.
(Slide change.)
MR. EMERSON: We agreed to develop the
implementing guidance for the rulemaking. With a
number of issues, the fact that this is a relatively
new area for rulemaking, adoption of a risk-informed
method, we're moving forward in parallel with the
staff in the development of this implementing
guidance. Some of the methods for resolving issues
that come up and making this an acceptable and a
useful rule, some of these might have to be resolved
in the rule language directly and these are issues
that we've discussed in the past. Some of them may be
resolved by putting information in the statements of
consideration for the rule and some of them can be
addressed in the implementing guidance.
So the implementing guidance is one of the
vehicles for addressing issues as they come up during
the next 15 months or 18 months, I guess.
The implementing guidance, we expect the
NRC will utilize in a Regulatory Guide once they have
agreed to it. The guidance is being developed by a
multi-discipline contractor team which addresses the
various areas of fire protection that need to be
addressed in a new rule. That includes classic fire
protection and safe shutdown, PSA, etcetera.
The rule language, as I said because the
rule language is being developed in parallel with the
implementing guidance, issues will come up and both
are going to be vehicles for resolving these
differences.
I think it's fairly safe to say that both
the industry and the staff are interested in coming up
with a clearly understood rule and with clearly
understood implementing guidance to support it and
we're looking forward to getting this in place to
risk-informed fire protection regulation.
Next slide, please.
(Slide change.)
MR. EMERSON: Let me just make sure I'm in
sync with the slides ont he screen. The fundamental
industry positions are the four that I've laid out on
the screen here and I'm going to spend each of the
next four slides, I'm going to elaborate on those a
little bit.
Next slide, please.
(Slide change.)
CHAIRMAN APOSTOLAKIS: So what -- you
didn't explain. Let's go back to 6.
MR. EMERSON: Back to 6, please.
(Slide change.)
CHAIRMAN APOSTOLAKIS: Licensees should be
able to use tools whether or not they transition to
NFPA 805. What does that mean?
MR. EMERSON: It means that the intent of
this rule was so that a licensee could choose an
alternate licensing basis.
As Eric pointed out, licensees have had a
licensing basis with which they've been comfortable
for the last 20 years or so in fire protection. And
if they choose, for whatever reason, not to adopt this
alternate licensing basis, we feel very strongly that
they should still have the ability to use
risk-informed performed-based tools in a structured
regulatory environment. So we would like to see the
methods that have been crafted over several years in
the Technical Committee available to licensees who
choose not to use a different licensing basis.
CHAIRMAN APOSTOLAKIS: The way the
standard is raised, the moment I use risk-informed
risk methods, I'm in NFPA 805.
MEMBER SHACK: You can use it as the basis
for an exemption.
MR. EMERSON: That's correct.
MEMBER SHACK: That would be the vehicle.
MEMBER POWERS: It seems like it should be
absolutely noncontroversial. If you want to use risk
bases to change something in your fire protection
program, and you're within -- you're currently in
Appendix R as your licensing basis, it's a perfectly
acceptable thing to do.
MEMBER ROSEN: But the extent is in 50.12
and it may be more difficult to do that if you have
805 out there.
MR. EMERSON: You can use risk tools now
to support exemption requests. What we don't have now
is a regulatory structure for processing that.
CHAIRMAN APOSTOLAKIS: But if I use this
method, don't I then have to show the CDF is less than
the value of -- how would that be different from using
NFPA 805? I don't understand.
MR. EMERSON: Again, the difference
between what we have now and what we're proposing for
a licensee that doesn't have -- that isn't going to
adopt the alternate licensing basis is he needs a
structure and we think the staff needs a structure too
for the acceptance of risk analyses to support
exemption requests. Right now the beauty is in the
eye of the beholder. I don't think -- I can't speak
for --
CHAIRMAN APOSTOLAKIS: My point is that if
the staff develops that kind of guidance it would be
an NFPA 805.
VICE CHAIRMAN BONACA: No, he's talking
about the criteria versus the application. What that
means is you can right now go in Appendix R that gives
you the criteria that you have to apply or 805 which
is a risk-informed criteria. I'm talking about -- it
says if somebody is using still Appendix R criteria
can use PRA --
CHAIRMAN APOSTOLAKIS: To do what?
VICE CHAIRMAN BONACA: Well, to make
changes from -- amendments, whatever to justify
changes to its own criteria.
MR. EMERSON: George raises a good point.
CHAIRMAN APOSTOLAKIS: The only way it can
be justified is to compare it to 1174 which means you
are using now 805.
MEMBER POWERS: George, 805 doesn't say
anything about this stuff.
CHAIRMAN APOSTOLAKIS: 805 says if you use
a risk method, you are using me.
MEMBER POWERS: No, you're not.
CHAIRMAN APOSTOLAKIS: That's what it
says.
MEMBER POWERS: No, it doesn't.
CHAIRMAN APOSTOLAKIS: What does it say?
MEMBER POWERS: I would love to see you
find that language that says --
CHAIRMAN APOSTOLAKIS: Tell me how it
would be different if I used risk methods to do it
within Appendix R that would be different from 805?
MEMBER POWERS: If I don't like something
in 805, I mean in Appendix R, nice prescriptive
regulation, I don't want to do that any more, I can
develop a probabilistic argument that says changing it
to something that I do like and I can develop a risk
basis on that and come to the staff and ask them to
approve it and never say a word about 805.
MR. EMERSON: George's concern is one that
--
CHAIRMAN APOSTOLAKIS: What I'm saying is
don't mention 805, but that's what you're doing.
MEMBER SIEBER: No, if you use 805, you
don't have to apply with an approved method. You
don't have to apply for the exemption. You just go
and do it, make a record and then once a year, like
50.59 changes, they come in and --
CHAIRMAN APOSTOLAKIS: The first time you
have to come to the staff.
MEMBER SIEBER: First time, yeah.
CHAIRMAN APOSTOLAKIS: And that's like the
first time what Dennis is talking about. All I'm
doing is I'm developing a risk argument. Then I have
to show --
MEMBER SHACK: No, ever time you want an
exception --
CHAIRMAN APOSTOLAKIS: That's the only
difference.
I can't imagine that's what they want.
MEMBER SIEBER: The basic rules for 805,
you know, separation, barriers and that kind of stuff,
where you have to end up after the fire is over are a
little different.
CHAIRMAN APOSTOLAKIS: Yes. Right now,
whether I like it or not, I comply with Appendix R
with exemptions. So the starting point is the same
for everybody.
MEMBER SIEBER: Right.
CHAIRMAN APOSTOLAKIS: I want to use a
risk argument now. I don't like it that they want me
to have this cable up for 10,000 feet. I'll do it
only for 6,000. I'm going to use a risk argument.
That means automatically I'll have to have some
baseline PRA, otherwise, I can't place it in context.
I have to use some model to calculate a difference in
risk from the 10,000 to 6,000.
MEMBER SIEBER: That's right.
CHAIRMAN APOSTOLAKIS: I have to have a
delta CDF. I'll have to argue that my
defense-in-depth is not suffering very much, right?
The usual arguments. And I don't mention 805.
MEMBER SIEBER: Right.
CHAIRMAN APOSTOLAKIS: Then I want to do
an 805. Tell me what I would do different? I would
do the same thing.
MEMBER SIEBER: You would do two things.
You would apply to use 805 in your plant and that
would probably be easily granted.
CHAIRMAN APOSTOLAKIS: If the methods have
already been approved.
MEMBER SIEBER: Well, they approve the
methods. You pick from among the methods they approve
and then for the very first time that you use it, you
say I'm using -- you send docket and say I'm using 805
to do this and this and this. Here's my --
CHAIRMAN APOSTOLAKIS: So for the very
first time, they're the same.
MEMBER SIEBER: Right.
CHAIRMAN APOSTOLAKIS: There's no
difference.
MEMBER SIEBER: There's another step for
805.
CHAIRMAN APOSTOLAKIS: Except for the
ultimate state.
MEMBER SIEBER: You have to get approval
to use it.
MR. EMERSON: This discussion points out
that the fact that there will be a spectrum of areas
where the licensee will fall into, ranging from full
Appendix R to full 805. And we, in developing our
implementing guidance are trying to structure it so
that wherever the licensee finds himself in that
spectrum, he has a consistent set of guidance so that
he can move farther along.
MEMBER POWERS: Let me understand. My
reading of what the staff said was that one is either
805 or one is not, that one can't go through and say
I'm 805 here. This part of 805 here and for the rest
of it I'm Appendix R.
MR. EMERSON: I would say in my view that
hasn't been completely worked out yet, how the partial
cases will be handled, whether you use a declaration.
MEMBER ROSEN: It seemed to be pretty
clear to me there's a moment in time when the licensee
sends a letter in that says I'm adopting 805. If you
don't have -- and the staff says okay, in a very
simple process. If you don't have such a letter,
you're not under 805. That's very clear.
MEMBER SIEBER: But the transfer --
MEMBER ROSEN: You can still use the
methods, but that may be endorsed in the 805 process,
but when you do and you want to make a change you have
to ask for an exemption.
MR. EMERSON: So there are certain
fundamental things you have to do to put yourself in
that camp.
MEMBER SIEBER: But to move from Appendix
R to 805 is supposed to be a bumpless transfer. In
other words, if you comply with Appendix R, you
automatically comply with 805 and you use 805 when you
want to make a change and you do that by area by fire
area. So you may have a plant that is 90 percent
Appendix R and 10 percent where you have modeled, 10
percent is the fire area that you have modeled.
CHAIRMAN APOSTOLAKIS: I don't think that
there is such a thing as complying with NFPA 805.
There is nothing to comply with. It tells you what to
do if you want to make some changes.
MEMBER POWERS: George, there are a bunch
of things in 805 you have to comply with.
CHAIRMAN APOSTOLAKIS: Like what?
MEMBER POWERS: Eric gave us a whole slide
of them.
CHAIRMAN APOSTOLAKIS: These are already
complied with.
I already do that.
MEMBER POWERS: And you're going to have
to comply tomorrow and the next week and the week
after that.
CHAIRMAN APOSTOLAKIS: Right.
MEMBER POWERS: And so there's a lot in
805 you have to comply with.
CHAIRMAN APOSTOLAKIS: If I say today, I
have my plant that's operating. I declare as of noon
today, I comply with NFPA 805, but I'm not going to do
anything to my fire protection, am I going to change
anything? No. Unless I decide to change something,
NFPA 805 doesn't do anything to me.
MS. BLACK: This is Suzanne Black.
Actually, 805 applies in all modes of the plant
operation, whereas Appendix R only applied for the
operating mode. So there are certain things that you
would have to consider in putting your fire protection
plan up front before you started to use 805.
CHAIRMAN APOSTOLAKIS: If I want to change
something.
MS. BLACK: If you want to change
something.
CHAIRMAN APOSTOLAKIS: If I don't want to
change anything, I don't have to change anything.
MEMBER POWERS: George, if you said I
don't want to comply with 805, I'm currently in
compliance with branch technical position and today I
declare I am 805, there are a lot of things we have to
do. One is that you have to do a site safety
assessment.
CHAIRMAN APOSTOLAKIS: If I don't plan to
go to the NRC and make some changes, I don't have to
do any of that.
MEMBER ROSEN: Yes, you do. If you're in
compliance with 805, you've done a site safety
assessment.
CHAIRMAN APOSTOLAKIS: I would be crazy to
say I'm in compliance with 805 when I already have
approval of Appendix R if I don't plan to change
anything.
MEMBER ROSEN: No, not crazy. I think
you'd be crazy like a fox, myself, because although
you will have to look at fire protection provisions
during shutdown which is the expansion that Suzie just
mentioned, you now have a world of flexibility to undo
the prescriptiveness of Appendix R where it doesn't
make any sense.
CHAIRMAN APOSTOLAKIS: Undoing means
changing something.
MEMBER ROSEN: Yes.
CHAIRMAN APOSTOLAKIS: And if I don't want
to change anything, nothing happens.
MEMBER ROSEN: You still have to --
MEMBER SHACK: You get an up front cost.
MEMBER ROSEN: Yes.
CHAIRMAN APOSTOLAKIS: Only if I want to
change something.
MEMBER SHACK: No, if you went with 805.
MEMBER POWERS: Why are you denying that
you have to do a site safety assessment?
CHAIRMAN APOSTOLAKIS: Because the whole
purpose of this, it seems to me, is to justify
changes.
MEMBER POWERS: No, it's not.
CHAIRMAN APOSTOLAKIS: Okay.
MR. EMERSON: You can also use it to put
yourself on a position to address future issues as
well without being subject to Appendix R directly.
CHAIRMAN APOSTOLAKIS: And future issues
again means changes, right? It might lead to changes.
MEMBER ROSEN: Revealed problems and then
you have some flexibility. Right now, you don't have
any.
It's not something you want to do by
volition, but the battleship in the desert is an
analogy which is you find something, you don't know
how the battleship got there, but you find something
and now the question is it acceptable.
CHAIRMAN APOSTOLAKIS: Right.
MEMBER ROSEN: And under Appendix R, it
may very well not be, but under this NFPA 805, you get
a chance to do a rational engineering analysis and so
it really doesn't matter.
CHAIRMAN APOSTOLAKIS: I agree with that,
because that's also change from where I am.
MEMBER WALLIS: Okay, so we'll agree there
is some change then.
CHAIRMAN APOSTOLAKIS: Huh?
MEMBER WALLIS: Can we move on, George?
CHAIRMAN APOSTOLAKIS: We can always move
on.
MEMBER ROSEN: Not if you, as the
Chairman, is asking questions.
CHAIRMAN APOSTOLAKIS: Oh, come up.
MEMBER ROSEN: Fred, let's see whether or
not George will ask questions.
Go ahead to your second bullet.
MR. EMERSON: Again, this slide indicates,
Slide 6 indicates the industry positions and if I can
work my way through the next four slides, I'll address
each of those.
Slide 7, please?
(Slide change.)
MR. EMERSON: The use of risk information
is a fundamental element. There's a fundamental
reason why we're supporting and participating in this
rulemaking is because we would like to see a lot more
ability to use in a structured and regulatory
environment, risk and performance tools.
We just talked about whether the tools
should be useful for all licensees or not and it's a
strong tenet of our position that they should be.
They should be able to avail themselves regardless of
what licensing basis they choose to put themselves
under. And a structured process for doing that needs
to be in place for both ends of the spectrum.
We see this as potentially an evolutionary
process where a licensee may choose to use 805 to
address certain specific issues that he's dealing with
at his plant and so he will change his licensing basis
to allow himself to address that specific issue, but
he may also see other issues down the road where he
finds it advantageous, so there may be a transition
process associated with his adoption of it and we
would like that to be a seamless process wherever he
chooses to place himself in that spectrum.
CHAIRMAN APOSTOLAKIS: This issue related
to what the Commission said some time ago that okay,
the risk-informed approach is an alternative,
voluntary alternative, but you can't just pick and
choose. You can't say here, I'm going to do risk,
here I'm going to do something else and it seems to me
that's what you're arguing here or how is it
different?
MR. EMERSON: There's been a lot of
discussion over what cherry picking is and whether
this constitutes cherry picking. On whether you
should adopt -- you could adopt 805 on a fire area by
fire area basis or on an exemption by exemption basis.
If you didn't want to adopt 805.
And there's the extent -- I think there's
a fair amount of agreement between the staff and the
industry now that partial use in some fashion is
acceptable and by partial use the industry has
consistently stated that we should be able to use the
tools as needed within the current environment, not
necessarily to change one fire area to be the 805 and
the rest of them be Appendix R.
MEMBER POWERS: This continues to be a
source of confusion to me because I read words that I
find acceptable which says 805 is an integrated whole.
Thou shalt not adopt it piecemeal.
But I keep seeing these words that say
partial use. If you're just talking about tools, I've
got no troubles about that, don't even know why it's
an issue, but that's between you and the staff.
Is it very clear that if you are 805, you
are 805? You're not Chapter 2 of 805 and something
else for everything else.
CHAIRMAN APOSTOLAKIS: Well, what he's
saying is yes. If you're in 805, you're in 805. But
he's also saying if you are in Appendix R, in some
parts of it you can be risk-informed.
MEMBER POWERS: And has that ever been a
question? I mean for the last four years has there
ever been a question about that?
CHAIRMAN APOSTOLAKIS: I don't know. Has
there?
MEMBER ROSEN: No. Absolutely not. Any
time you want to ask the staff for an exemption from
Appendix R requirements, you can. And 900 cases of it
are --
CHAIRMAN APOSTOLAKIS: Using risk
information?
MEMBER POWERS: Nine hundred cases of
them, no risk information was ever used. Most of them
are
-- I just can't -- it doesn't fit.
CHAIRMAN APOSTOLAKIS: Can you use risk
information for this little fire area and for the rest
of the plant Appendix R?
MEMBER ROSEN: If you go through the
exemption process, I think so.
CHAIRMAN APOSTOLAKIS: But the exemption
process doesn't allow risk? Does it allow for risk?
MEMBER SIEBER: Sure.
MR. WEISS: This is Eric Weiss of the
staff. The staff is prepared to entertain an
exemption on any basis, whether it's risk or
performance-based or whatever. And that's not to
prejudice the outcome of the review. We can't say
simply because one puts risk at the top of the page
that the exemption is going to be granted, but I'm
sure the staff would entertain an exemption on any
basis and give it careful consideration.
MEMBER POWERS: And if the basis came in
on risk they have a regulatory guide to help them
assess that. It's one you're reasonably familiar
with.
CHAIRMAN APOSTOLAKIS: And my point is
that if that's the route you want to take, you're in
805.
MEMBER POWERS: No.
CHAIRMAN APOSTOLAKIS: Okay.
MEMBER SHACK: Can you explain to us --
I'm sort of perplexed like Dana, it seems to me. The
tool is available. Sure, they're available. What is
the point of contention here? That's what I'm
missing.
MR. EMERSON: The point of contention was
when we first started talking about rulemaking to
adopt NFPA 805, there was some discussion as to
whether partial use was allowed, whether you could --
a licensee who chose not to adopt the standard could
make any use whatsoever of the tools and from the
beginning we've been proposing that the licensee who
chooses to maintain his existing licensing basis
should not be shut out from the use of the tools that
have been crafted in NFPA 805 just because he chooses
not to adopt it as an alternate licensing basis.
So what we've been working on is ways that
he can use those tools on a structured environment.
CHAIRMAN APOSTOLAKIS: I guess it's too --
MEMBER SHACK: The magic words, structured
environment, I guess is the point of contention.
MR. EMERSON: So if a licensee chooses not
to adopt 805, but he sees a use for risk tools to
support an exemption request, that he understands the
bounds on his use of the tools and the acceptable
nature of his use of the tools when he submits that
request and that should be consistent with the way the
tools are used by the licensee who does choose to go
to 805. So that at some point if he chooses to make
a transition, the process, if he's used are consistent
from one to the other.
MEMBER ROSEN: Fred, we've been keeping
you from moving forward, but let's just see what we
can do in the next few minutes.
MR. EMERSON: Okay. Slide 8, please?
(Slide change.)
MR. EMERSON: This really isn't an issue.
It's been a fundamental tenet from the beginning of
our discussion of 805, but it's never been an issue
with the staff who have always agreed with this
contention.
Slide 9, please.
(Slide change.)
MR. EMERSON: The transition process is
obviously of great interest to any licensee who is
going to be contemplating changing licensing bases.
The transition needs to be relatively uncomplicated or
as uncomplicated we can make it and still maintain an
acceptable regulatory process.
The mere fact of a transition does not
either make the licensee more safe or less safe. And
all that means is that he has a different regulatory
environment in which to consider changes to his plan.
It's critical that the process, the
transition process be well understood by both the
licensee and the staff. The licensee has a clear idea
of where his licensing basis is which I'll touch on in
a minute throughout the transition process. The
licensee knows what he has to submit and what he can
retain; when a license amendment is required and when
it isn't. All those sorts of things need to be
addressed very clearly.
Now uncomplicated doesn't necessarily mean
easy. And we would expect on the industry side that
a licensee would have to do a fair amount of work to
identify his current licensing basis and how he stacks
up against provisions of 805 that he would either like
to take advantage of or bring forward an alternate
approach for, but nonetheless, it will be a fair
amount of work for the licensee to put himself in 805
space. What he has to submit may not reflect the
amount of work he has to do, but we want him to be
thoroughly prepared.
Slide 10, please.
(Slide change.)
CHAIRMAN APOSTOLAKIS: Even after the IEEE
work, there would be a lot of work?
MR. EMERSON: Well, as I say, the licensee
needs to -- he needs to have a thorough understanding
of his current licensing basis and he has to have a
thorough understanding of what alternate approaches he
might have to make to adopt 805. For instance, he has
to consider all of the fundamental elements of Chapter
that Eric had on his slide. He wants to know how his
current licensing basis stacks up against each of
those elements. He may choose to say okay, I can
agree with what Chapter 3 says so in this particular
-- in one particular -- for one fundamental element he
may say Chapter 3 is okay. For another, he may say I
have my own licensing basis well established that I'd
like to bring forward in place of this and that's also
allowed by the standard. Or he may have a third
method that he'd like to propose in which case the
staff has to review it, but he needs to consider all
of those things to see how -- what the level of effort
will be for him to move to a different licensing
basis.
CHAIRMAN APOSTOLAKIS: Well, but what I'm
saying is that the fact that the licensees have
already done an IPEEE would be a significant --
MR. EMERSON: That's a significant help,
yes.
MEMBER POWERS: Do you really think so?
It seems to me that when we went through the
functional fire protection inspections that licensees
were spending on the order of a million just to get
their licensing bases in order, even though they had
done an IPEEE.
MR. EMERSON: All I'm saying is that --
CHAIRMAN APOSTOLAKIS: It works if they
hadn't done the IPEEE.
MEMBER POWERS: I don't --
CHAIRMAN APOSTOLAKIS: The first step is
always --
MEMBER POWERS: I don't think the IPEEEs
that were done were either unduly laborious or very
helpful.
CHAIRMAN APOSTOLAKIS: Either what?
MEMBER POWERS: Unduly laborious or very
helpful.
MEMBER ROSEN: I think that's besides the
point.
MEMBER POWERS: Yeah, it's the thorough
understanding of the licensing basis and that's a big
job.
MR. EMERSON: One of the most important
points of the transition is that the licensing basis
has to be clearly understood. If you're talking about
partial implementation in any form, if you're just
talking about selected use of it, if you're talking
about a transition process that changes through time
--
MEMBER POWERS: You keep choosing this
word that I just don't understand.
MR. EMERSON: What's that?
MEMBER POWERS: This "partial
implementation".
MR. EMERSON: Let me --
MEMBER POWERS: I think you need to say
805 is or it is not.
It's an integrated whole. You can't use
half of it.
MR. EMERSON: You may choose to put
yourself under the rule with 805. You may not use it
in all areas of your plant immediately. You may use
it initially for changes in only a few areas of your
plant. But you have placed yourself under 805. Both
the licensee and the staff need to understand where
you've chosen to apply that throughout the process,
since it won't be an instantaneous transition.
CHAIRMAN APOSTOLAKIS: Isn't that what
that big box with deterministic and probabilistic --
I mean that's what it does.
MR. EMERSON: That's the process of
analyzing a change once you have placed yourself in
this area.
CHAIRMAN APOSTOLAKIS: Right.
MR. EMERSON: So that when you implement
a change, you can select a technique.
CHAIRMAN APOSTOLAKIS: Is there a good
example somewhere with specifics as to what exactly
you mean? I will not be in 805, but I'm allowed to
use the tools? I don't understand that.
MR. EMERSON: Well, the example is if you
wanted an exemption request to address a fire barrier
so you discovered your fire barrier and you're in a
certain fire area was not what you thought it was --
CHAIRMAN APOSTOLAKIS: Right.
MR. EMERSON: But you thought you had good
grounds for leaving it the way it was instead of
making a repair, you could -- or making a design
modification, you could utilize 805 tools to support
an exemption request instead of making the design
change.
CHAIRMAN APOSTOLAKIS: Okay, so that
exemption request now would require me to calculate a
delta CDF, would it not?
MR. EMERSON: It may well require that.
CHAIRMAN APOSTOLAKIS: Well, there's no
other way.
MEMBER POWERS: Yes, there is.
CHAIRMAN APOSTOLAKIS: Like?
MEMBER POWERS: Any way -- on the same
basis that 960 exemption --
CHAIRMAN APOSTOLAKIS: Then I'm not using
risk information.
MEMBER POWERS: It didn't say anything
about using risk information.
CHAIRMAN APOSTOLAKIS: That's what he
says, the tools are risk tools.
MEMBER ROSEN: Not necessarily. There's
fire modeling. You can model the effect of the fire.
It's not a risk analysis. It's an engineering
analysis.
CHAIRMAN APOSTOLAKIS: That's what you
mean?
MR. EMERSON: That's one possibility you
could use risk or you could use fire modeling.
CHAIRMAN APOSTOLAKIS: Those tools are not
used in Appendix R?
MEMBER ROSEN: Fire modeling, no.
CHAIRMAN APOSTOLAKIS: Okay, now I've got
my example. But the moment you get into risk though,
it's a different ballgame.
If by tools you mean some code that
calculates thermal fluxes, okay, fine.
MEMBER ROSEN: That's just one of the
tools.
MEMBER SIEBER: But then it's a risk
analysis, another tool, right? But used outside of
805 for an ordinary submittal for an exception.
CHAIRMAN APOSTOLAKIS: I disagree.
MEMBER ROSEN: What's where you disagree,
but the risk analysis simply says yeah, there could be
a big fire here and it could burn everything down, but
it doesn't matter, here's why.
The stuff that burns down is not risk
significant. It doesn't help me get the safe
shutdown, either hot or cold. That's typical of a
risk analysis.
It's very complicated. Please go ahead.
MR. EMERSON: Again, the fundamental point
here is that the licensee and the NRC both need to
understand what the licensing basis is throughout the
transition process.
Slide 11, please.
(Slide change.)
CHAIRMAN APOSTOLAKIS: But you've got to
play devil's advocate now. Are you saying that if I
don't want the transition, I don't have to have a
thorough understanding of my licensing basis?
MR. EMERSON: No, I didn't say that.
CHAIRMAN APOSTOLAKIS: You didn't say
that, okay.
MR. EMERSON: The licensee should --
MEMBER POWERS: But the truth of the
matter is that based on the sampling of four plants
that right now it is a struggle to coil together what
the current licensing basis is because in many cases
this is now a 21-year old assembly of information.
MEMBER ROSEN: But Dana, some plants have
invested in design basis documents and have their fire
protection design basis well documented.
MEMBER POWERS: Well, I know four of them
that do. And you're hinting to me there may be a
fifth, but I know four that definitely got theirs in
order. But I also know it took a substantial effort
to do that.
MEMBER ROSEN: It did.
MR. EMERSON: I'll try to move through the
rest of the slides fairly quickly.
Slide 11, please.
(Slide change.)
MR. EMERSON: The current rule language as
the staff indicated, they have put several drafts of
the rule language, made them available on the web to
stakeholders and we've had several interactions with
the staff to discuss these various drafts. The
current draft was put out, I think last week. It's
the third draft that's been made available. We have
some positive comments about it. We have some
concerns about it, but these concerns, I would like to
downplay a bit because I think these will be addressed
as we move forward with the staff, them developing the
rule, us developing the guidance and we'll get these
addressed.
Slide 12, please.
(Slide change.)
MR. EMERSON: The implementing guidance is
being written as we speak. I'd only like to emphasize
-- I don't have a draft to share with you yet. I'd
just like to emphasize quickly the key elements of it.
The main body of the implementing guidance is how to
go through the process of making a transition. The
key elements of that are what the process is, what the
options the licensee has, what guidance there may be
for a licensee choosing to maintain his existing
licensing basis with the possible use of 805 tools or
guidance for how to adopt a new licensing basis. And
then lastly, how to maintain configuration control
over his licensing basis once he has made the
transition.
Slide 13, please?
(Slide change.)
MR. EMERSON: The appendices in the
implementing guidance are basically how one looks at,
interprets or uses the provisions of NFPA 805 itself,
how you do the -- how you look at the program
fundamentals, how you establish performance criteria,
identify fire hazards in your systems structures and
components and how you would do an evaluation against
the performance criteria.
Overall, our schedule for completing the
implementing guidance is in parallel with the rule.
As I said, we'll have a first draft of the NRC staff
later this month. We'll have a second draft after the
Commission has issued their instructions to the staff,
maybe later this fall. And the third draft in the
spring.
Slide 14, please.
(Slide change.)
MR. EMERSON: The next several slides I've
identified some potential barriers, hurdles. These
all fall into the category of things that need to be
considered and worked out as we move forward with
developing the rule language and the guidance. They
need to be addressed, as I said, either in the
language, in the statements of consideration or in the
implementing guidance.
One of the areas is where a license
amendment may or may not be required for a submittal,
for either an initial or subsequent submittal. One is
the definition and use of the current licensing basis,
if one makes the transition, to what extent can the
licensee bring forward elements of his current
licensing basis to replace elements of 805. Those are
several technicals issues. We've had on-going
discussions with the staff on exemptions -- I'm sorry,
exceptions to the rule that may or may not --
exceptions in the rule to the standard that may or may
not be necessary and again, we're continuing to
discuss these with the staff.
Slide 15.
(Slide change.)
MR. EMERSON: Because 805, I'm using 805,
requires some new elements that are not currently part
of Appendix R, areas like monitoring and shutdown and
low power modes, we classify these as potential
hurdles, only because we haven't dealt with them
before and we will have to be able to do that
successfully to create a successful rule and matching
implementing guidance.
Slide 16, please.
(Slide change.)
MR. EMERSON: The benefits we see, we see
an allowance for the use of risk methods and resolving
current fire protection issues that neither plant
specific or generic issues that the licensee may be
confronting. We see it as being able to address the
four NRC organizational goals or pillars of
maintaining safety and increasing public confidence
and reducing regulatory burden and increasing
efficiency and effectiveness. We see this as being
able to address all of those.
We see it being able to focus fire
protection programs on things that are more
risk-significant. Right now, we don't have that
capability under Appendix R, applying resources where
they make the most sense. We see it as providing a
consistent method for supporting exemptions,
deviations, 50.59 and 86-10 evaluations and we see it
as providing, as requiring a seamless transition
process from the deterministic to the risk-informed
regulatory framework.
Also, we see the use of risk methods. If
you integrate a consideration of fire risk into
overall plant risk that will help us resolve issues
for fire protection interests and other plant
operating interests are competing and it allows us to
evaluate both on an equal basis, on an equal risk
basis.
Slide 17, please.
(Slide change.)
MR. EMERSON: The subcommittee, when I
spoke on Tuesday asked me to hazard a guess as to how
many licensees might adopt this rule, so what I'd like
to do is to lay out a scenario and this is really what
I think the way it will unfold and how licensees will
consider and use this.
First, we -- I see that we need to
complete the current efforts that are on-going to
improve our ability to use fire risk. If the ANS fire
PSA standard development, I think that will be a
useful fundamental point to buttress the use of risk
information and I think the EPRI and research effort
to requantify fire PRAs and improve our ability to use
fire risk techniques will also help there.
Secondly, I see a few plants and we saw an
example during the subcommittee meeting of one
utility, at least one, not the only that would be
considering using this rulemaking. The plants that
are most likely to look upon this favorably are ones
that are used to using risk techniques in their normal
plant operations. And have established PRAs that they
can and have relied on. Those will be the plants that
are most likely to adopt this first. I see them using
the tools and the 805 basis for successfully and in a
few evaluations and then continuing to expand their
use of it as they have successful regulatory
applications.
Slide 18, please.
(Slide change.)
MR. EMERSON: Once the rest of the
industry sees successful use and successful regulatory
interactions, using risk tools in the fire area, I
think more and more plants will move to adopt this and
they'll see the benefits. The benefits, I think, will
be too obvious to ignore and I think you'll see plants
improving their own risk tools and their ability to
use them in this structured environment the 805 will
afford.
So again, I eventually expect to see most
plants using this to some degree and I put a big if
there.
Slide 19, please.
(Slide change.)
MR. EMERSON: I think it depends a lot on
what the staff and the industry can accomplish in the
next 15 or so months as we develop the rule language
and the implementing guidance. We can make this a
clear rule, a useful rule and a rule that's soundly
supported by clearly understood guidance. If we're
not successful, we can create a rule that's more
difficult to apply and is less attractive to someone
who is considering taking advantage of it. So that's
the challenge ahead of us now is to create an
effective set of combinations --
MEMBER WALLIS: Are these generic
statements or have you identified barriers and
hurdles?
MR. EMERSON: There are some barriers and
hurdles that I listed on the previous slides.
MEMBER WALLIS: Yes, those are the ones
you listed before?
MR. EMERSON: I'm sure there will be
things that come up that we haven't foreseen.
MEMBER POWERS: The problem I have is when
I compare your lower hurdles to your list of hurdles,
one of those hurdles was dealing with the shutdown
mode of operation. Don't you mean surmount hurdles?
MR. EMERSON: That might be a more
accurate way to state it.
MEMBER POWERS: I don't think you want to
say oh, well, let's just take the shut down
requirements out of 805.
MR. EMERSON: No, that's not what I
intended.
That completes my presentation.
MEMBER ROSEN: Does the Committee have any
further comments on this subject?
MEMBER POWERS: Yeah, are we discuss
associated circuits?
MEMBER ROSEN: Well, we didn't intend to
do that.
Dana asked whether we were going to
discuss associated circuits, the NEI document, the
discussion the subcommittee had on that, NEI 00-01 and
we did not intend to go into that today. It's not as
fully far along, Dana, as 805. There is clearly a
link between the two. Ultimately, I would hope that
NEI --
MEMBER POWERS: Well, where do we stand?
MEMBER ROSEN: Will, in fact, be one of
the methods adopted by the Regulatory Guide and that's
the linkage.
MEMBER POWERS: Where does the staff stand
on associated circuit analysis? We've suspended
inspecting on it. Are we still in suspension?
MR. HANNON: Yes, this is John Hannon. We
currently have the hiatus on the inspection in place
and are looking to resume inspection some time in the
March time frame of 2003, given we can reach a
consensus on the appropriate approach including the
risk-informed aspects with the stakeholders.
MEMBER ROSEN: Are there any other
Committee questions? If not, seeing none, I turn it
back to you, Mr. Chairman.
CHAIRMAN APOSTOLAKIS: We didn't really
discuss the standard itself, did we, the contents? It
was all process stuff.
MEMBER ROSEN: I think we discussed it in
general terms, but the --
CHAIRMAN APOSTOLAKIS: But what it exactly
does was not discussed.
MEMBER POWERS: Well, we've been through
it.
CHAIRMAN APOSTOLAKIS: Two years ago when
we rejected it.
MEMBER POWERS: We didn't reject it. We
just said we didn't anticipate that many people would
work with it. Now we're told that eventually all of
them are going to do it.
CHAIRMAN APOSTOLAKIS: It was not
risk-informed at the time.
Anyway we'll break until 10:30.
Yes? Certainly, come in.
MR. GUNTER: Paul Gunter, Nuclear
Information Resource Service. The issue of public
confidence, I think, is what I'd like to speak to
because as an onlooker into the staff meetings on this
process, I think that -- as well as the ACR meeting,
I think that we get a sense that -- of a clear warning
that there's going to be a lot, even more problems in
context when this enters into the inspection and more
importantly the enforcement process. I can't help but
think that what we are seeing is, in fact, another
overlay to a very complicated labyrinth that we've
seen since the original introduction or even the fight
over the introduction of Appendix R and now we're
about to have another overlay on this whole
complicated, but very significant contributor to core
damage frequency. And the public confidence is really
looking for areas where we can trust that there's
going to be enforcement. And frankly, we don't see
that happening right now. And more particularly, our
concern is that this is just going to confound the
whole inspection process.
I guess what I would like to get some
sense of from staff is how -- just to speak to the
issue of how this is going to make the inspection
process more efficient and more particularly how we
can get out of a limbo of argument and contest to
effective enforcement.
CHAIRMAN APOSTOLAKIS: Any reaction from
the staff?
MR. WEISS: Yes, this is Eric Weiss. We
recognize Mr. Gunter's concerns. Let me say this. We
know that we have a challenge in front of us, but we
also a vision for how this will work and we recognize
that training is needed. We're conducting quarterly
training of inspectors now. That's one of the things
we've instituted in our branch. We're prepared to
give the inspectors tools and training in this area
and without going into a lot of detail, let me say
part of the vision for how this would work is that
inspectors are good at certain things. They're good
at determining whether approved methods have been
used, whether the people that use them are qualified
to use them, the inspectors are good at determining
whether the configuration that was in the design was
implemented properly in the plant and all of those
things have yielded big improvements in safety and
maintaining safety.
And we're prepared to meet that challenge.
We think that 805 will make the process more efficient
when we can construct a set of inspection criteria
that tied to those types of things.
Conversely, I can see that if we do a bad
job, if we ask the inspectors to go out and duplicate
the analysis that's being done in fire protection,
that would not be efficient. We recognize a challenge
and I think we can make it more efficient by
constructing the inspection process appropriately.
CHAIRMAN APOSTOLAKIS: Any further
comments? Okay, we'll recess until 10:35.
(Off the record.)
CHAIRMAN APOSTOLAKIS: Next item is
generic resolution of voids in the concrete
containment and this is under the strong leadership of
Dr. Bonaca.
VICE CHAIRMAN BONACA: Thank you so much,
Mr. Chairman. In March 2002, the subcommittee on
License Renewal in Florida City to review the SER of
Turkey Point application for license renewal. During
that meeting, a member of the public expressed
concerns with voids identified in the containment
walls of Turkey Point during the steam generator
replacement in the 1980s.
At the meeting, we requested that during
the April, full ACRS meeting, the staff presented to
ACRS the following: (1) if and how the issue was
resolved at Turkey Point; and second, how the licensee
and the staff had addressed the generic implications
of Turkey Point findings including communications or
generic implications at the other plants.
At the April meeting, the staff and
licensee made a convincing case that the issue was
properly resolved at Turkey Point. However, they
didn't have sufficient time apparently to find how the
issue was communicated or addressed generically and
they asked for more time to provide this information.
Yesterday, we received a memo from Gene
Imbro, the Division of Engineering, who is here. You
have a copy of that memo, which I believe convincingly
provides, first of all, the tracing of the
communications that took place and also addressed the
issue of the fact that there isn't a generic concern
with the void in containment. I will let you -- I'll
now leave the meeting to Gene Imbro. He'll give us a
presentation on these issues.
MR. IMBRO: Thank you. It's a pleasure to
be here to address you and hopefully resolve this
issue, provide you with a little bit of background.
I'm the Chief of the Mechanical Engineering and
Structural Branch in NRR. With me is Kamal Manoly,
he's a Section Chief in the Structural Section and we
wanted to talk to you, as Mario said, about assessing
the generic applicability of the construction defects
that were found at Turkey Point during the steam
generator replacement activities during 1982.
First slide.
(Slide change.)
MR. IMBRO: Just by way of background and
maybe a little refresher for us all, there are, of
course, regulatory measures in place to look at
construction defects and actually to control
construction. I mean first of all there's the
licensees' QA and QC program with complies with
Appendix B and which includes, of course, written
procedures and process to identify conditions adverse
to quality and corrective actions.
In addition to that, is also a reporting
requirement in 50.55(e), 10 CFR 50.55(e) where the
holder of the construction permit is required to
report construction deficiencies that would create a
substantial safety hazard. And superimposed above all
of this is the NRC's construction inspection program
that was identified or outlined in our Inspection
Manual, Chapter 2512 and for the later plants, of
course, this included a construction resident. There
was direct observation of construction activities
probably during -- for whatever vintage plant we're
talking about and they looked at evaluation of the
licensee and contractor performance and they evaluated
the licensee's control over the activities and of
course they talked to people involved with the
process. So it's kind of a multi-layer control here
of construction activities that we just wanted to
point out.
To respond to the issue at hand in terms
of whether or not the defect found at Turkey Point had
generic applicability we started out to do --
basically revisited what was done by the region, at
least in part, back during the 1980s, early 1980s time
frame of the steam generator replacement. We did a
search of NUDOCS dating back as it says on the slide
to 1968. We looked at a lot of 50.55(e)s. We
identified nine areas where defects were found in
concrete structures by the QA/QC program and were, in
fact, corrected. So I think this provides us with a
level of confidence that each QA/QC program in terms
of identifying issues and correcting them and the NRC
oversight was effective in providing some confidence
that these structures do not have significant voids.
Just as a little bit of a background,
also, most of the voids that were found were located
in congested areas of rebar and that would be
typically around penetrations or where the base mat
joins the containment shell.
MEMBER LEITCH: Did you say that nine
voids were actually reported or were these
deficiencies in the program that QA found?
MR. IMBRO: No, nine voids were actually
reported, nine different instances of voids were
reported and these were corrected.
VICE CHAIRMAN BONACA: Some of them during
construction?
MR. IMBRO: Yes.
VICE CHAIRMAN BONACA: And some of them
after construction, for example, Turkey Point?
MR. IMBRO: Yes, exactly, exactly.
VICE CHAIRMAN BONACA: Was accidentally
found.
MR. IMBRO: Yes.
VICE CHAIRMAN BONACA: Because they
changed --
MR. IMBRO: The steam generator.
VICE CHAIRMAN BONACA: Hatch opening.
MR. IMBRO: Yes.
VICE CHAIRMAN BONACA: So now we see what
happens when they're doing that at Davis-Besse.
(Laughter.)
MR. IMBRO: Good point.
VICE CHAIRMAN BONACA: It's also backed
out containment, isn't it?
MEMBER SIEBER: Different kind of
containment at Davis-Besse.
MR. IMBRO: Yeah, I'm not sure who's the
A/E, but I'll take your word that it's Bechtel.
MEMBER SIEBER: Davis-Besse is a thermos
bottle type containment so the strength of it comes
from the liner which is pretty thick there.
MR. IMBRO: Okay. In addition to the
programmatic activities that I just described, there's
also for containment a structural integrity test
performed and the structural integrity test is
basically to provide assurance, additional assurance
that the performance of the containment conforms with
the analysis, so what the structural integrity test
does is they pressurize a containment to 15 percent
above design pressure. The rebar is instrumented.
The instrumentation is concentrated primarily around
the areas where there are discontinuities in the
containment, for example, around penetrations or like
the equipment hatch, for example, and other areas
where there's a discontinuity in structure, either
because of an opening or a change in thickness and
they measure at the point where they pressurize the
containment to 15 percent above design pressure, they
measure the strain in the rebar and they compare it
against the A/E analysis. And if the analysis and the
strain measurements are reasonably close, then that's
an additional point of confidence.
For Turkey Point, I think this was --
Turkey Point did have a structural integrity test. It
passed the structural integrity test. So the rebar,
measured strains of the rebar were pretty much as
predicted and that would indicate that first of all
the void was relatively small and there was no issue,
excuse me, no issue with the containment performance
or the containment being able to perform under design
conditions at design pressure.
MEMBER POWERS: Do we have a data base?
It says okay, we run these strength tests and it's
been done with voids of various sizes so that we know
what the effect of voids of various sizes are.
MR. IMBRO: I don't know that answer.
MR. MANOLY: This is Kamal Manoly. I
don't know of any data base, in particular. It's
something that utilities or A/Es do following the
design to ensure the range between the design values
and the measured values are very close, not so much
that it has to be way below, but within the
calculation of accuracy.
You don't assume these voids when you
design concrete.
MR. IMBRO: Well, I think in direct answer
to your question, I mean the answer is I think no.
I'm not 100 percent sure on that. I don't think the
staff has ever done an evaluation to determine the
effect of the size of the void on the variations and
strain.
MEMBER POWERS: Well, I guess the
contention is well, the voids must be small, because
the tests came out so well and it's not clear to me
that the tests test for voids.
MR. MANOLY: You're testing for behavior
of the structure at design and you assume a
homogeneous structure, so if you do have major
discontinuities the cross section will behave
differently.
MEMBER SHACK: Yes, but it would be nice
if somebody put a void in the analysis, did the
analysis and said --
MR. IMBRO: Actually, in fact, please, go
ahead, I'm sorry.
MEMBER SIEBER: My way of looking at it is
a little bit different. I don't think the structural
integrity test tells you very much about voids because
when you pressurize the containment, basically all the
rebar and the concrete and everything else goes under
tension.
MR. MANOLY: Yes, but in that penetration
you get a lot of bending.
MEMBER SIEBER: Bending intention, but the
concrete cracks. And so -- and concrete isn't good in
tension. It's great in compression.
So if there's a void there, it's the rebar
that's holding the containment together because it
does perform well --
MR. IMBRO: Absolutely.
MEMBER SIEBER: -- tension and bending.
MR. IMBRO: The other point I think I
wanted to make too is particularly with respect to
Turkey Point. Bechtel at the time that this
construction defect was discovered, did reanalyze with
the void in place and they found that even with the
void in place that none of the allowables were
exceeded and the deflections were all --
MEMBER SIEBER: I would expect that.
MR. IMBRO: I don't know if I could add
anything more to that.
Go to the next slide.
(Slide change.)
MR. IMBRO: Recognizing that this -- all
this occurred in 1982 or thereabouts, some 20 years
ago, there was -- we did search our records and we did
find that there was an LER written by the licensee at
the time of the steam generator replacement which
identified the voids for the containment that were
found both for Units 3 and 4.
This was evaluated by the region at that
time, although none of the specifics are provided in
the report as to what they did, but they did consider
whether this had generic applicability. And since
there was no further generic action taken, I think
it's reasonable to conclude they felt there was no
generic applicability and I guess a conjecture, maybe
there are a couple of reasons. First of all, the void
was quite small, relatively speaking. And also, the
fact that the Bechtel analysis demonstrated that the
stresses even with the void as found didn't invalidate
the analysis or cause allowables to be exceeded.
So there was a trail that was the LER was
written. The region did follow up on the LER. The
LER was closed out and a regional inspection report
and the inspection report by our indication says
specifically that this LER was evaluated for generic
applicability.
MEMBER POWERS: When we do analyses for
severe accident phenomena, things like direct
containment heating, we construct a distribution of
loads on the containment and then we construct a
fragility curve for the containment. I think what
you've discussed here speaks to the issues of
containment performance at its design level. Would
any of these things, voids and what not, affect this
fragility curve that looks at the performance of the
containment well above the design level?
MR. IMBRO: Well, I mean, I think the
answer is yes. To what degree, I can't really cite.
Maybe that's something that may need to be
investigated at some point, the fact that there is a
potential for voids in containment albeit at least not
causing an issue at a normal design condition for for
severe accident conditions, yes. Maybe that's
something that needs to be looked at.
MR. MANOLY: But only, I think, if you
have major significant voids and we just -- I don't
believe that there is major significant voids in
structures that have seen all this programmatic
activity because you've got several layers of
oversight. So we have some voids, but I don't think
you're going to have anything really --
MR. IMBRO: Well, I mean, I think Mr.
Powers' point, Dr. Powers' point is a valid point,
that yes, there are voids, even though they may be
some relatively small size. That still should be
factored in or could be factored into the severe
accident fragility curves.
MEMBER POWERS: It's pretty clear we know
that little voids must not affect those curves very
much because we do experiments on model concrete
containments. They undoubtedly have little voids when
they place the concrete. I don't think they do
anything really that dramatically different from what
they do in reactors. So little ones, we know, are
already built into our experimental data base, so it
is a bigger void that you'd be worried about.
The question that comes back to me next
with is what's a bigger void and I don't know any
better than you do.
MR. IMBRO: This is a little bit beyond my
area. I'm not very knowledgeable in this part of this
--
VICE CHAIRMAN BONACA: If I remember, the
one at Turkey Point, there were 9 -- how many cubic
feet?
MR. IMBRO: The void was -- I think it was
at its widest point there was 9 feet wide. It went
the whole length of the -- the thickness of the wall
and it was -- varied in size from I guess a maximum of
about 17 inches to 6 inches. So there was -- I could
do the arithmetic, I guess.
VICE CHAIRMAN BONACA: That's an area
where the thickness --
MR. IMBRO: Maybe something like 20 cubic
feet or something --
VICE CHAIRMAN BONACA: Then the thickness
of the wall there is probably --
MR. IMBRO: It's about 8 feet, 7.5 feet I
think is the number.
MEMBER SIEBER: One of the issues that
might or might not be important is that concrete, even
though its contribution to the strength is not as much
as the rebar, its shielding effect is pretty
substantial.
MR. IMBRO: Yes, of course.
MEMBER SIEBER: And I was involved, not
responsible for it, but involved in a project where we
built tanks that looked like containments to store
really hot resin in and you could map where the voids
were.
(Laughter.)
MEMBER POWERS: In fact, that's a common
way to look at voids of anything is just to zap it
with gamma rates.
MEMBER SIEBER: That's right. And it
really showed up. I mean you could draw them out, but
in case of an accident where the concrete is missing
and all you have is liner and rebar, the radiation
field on the outside may be substantial which could
impact, depending on where it is, some operations, for
example in say the aux. building or penetrations --
has anybody considered that effect.
MR. IMBRO: No. We haven't. I'm not sure
that that was considered.
MEMBER SIEBER: Okay.
MR. IMBRO: Just as a quick wrap up, we
believe that there are substantial programmatic
controls out there that would prevent large voids from
forming, first of all, and then if they did, they
would be detected from several means. And I think
that our conclusion after revisiting again, well,
first to back track a little bit. The Region did
evaluate this back in 1982. There's a documentation
that via the LER and the Bechtel analysis and all
those things that were present at that time, probably
supported the Region's conclusion that this wasn't an
issue that needed to be pursued generically. We've
looked at it again now with new eyes, again 20 years
after the fact and we've reached the same conclusion
that we really think that this issue does not need to
have any generic further look.
MR. MANOLY: One thing I'd like to add
also in the SIT, structural integrity test, you do
mapping of the cracking. So if we have big areas of
voids, the pattern of cracking will change.
MEMBER SIEBER: Yes, it does.
MR. IMBRO: Any further questions?
MEMBER RANSOM: What were the licensee's
correction action that you mentioned here?
MR. IMBRO: The licensee, when the voids
were discovered, they excavated the concrete down to
solid concrete at Turkey Point and they regrouted it
again, so they replaced the concrete that they found
was missing. They excavated the solid concrete and
pumped fresh concrete in to fill in.
VICE CHAIRMAN BONACA: Then they exchanged
some in the next containment and they found voids in
the same location. That's why we asked those
questions about genetic implications.
MR. IMBRO: That's right. Okay, if there
are no further questions, thank you.
VICE CHAIRMAN BONACA: Any additional
questions? I think we've received answers as far as
tracing the fact that there was an LER issue and there
was a review performed by the region and I think that
we got the information we needed.
Any additional questions from Members?
MEMBER SIEBER: I just would comment that
it was a pretty comprehensive report. I was glad to
see that.
MR. IMBRO: Thank you.
VICE CHAIRMAN BONACA: Yeah, it's a nice
summary.
Thank you very much for the presentation,
well informed and thank you for the search and Mr.
Chairman, I'll turn the meeting back to you.
CHAIRMAN APOSTOLAKIS: Thank you very
much.
VICE CHAIRMAN BONACA: You have an
additional 45 minutes.
CHAIRMAN APOSTOLAKIS: Yes, we very much
appreciate that. I think we're ahead of schedule now
almost.
Okay, the next business is ACRS activities
and so on, so I don't think we're going to need a
transcript any more.
(Whereupon, at 10:58 a.m., the proceedings
were concluded.)
Page Last Reviewed/Updated Monday, July 18, 2016