488th Meeting - December 6, 2001
Official Transcript of Proceedings
NUCLEAR REGULATORY COMMISSION
Title: Advisory Committee on Reactor Safeguards
488th Meeting
Docket Number: (not applicable)
Location: Rockville, Maryland
Date: Thursday, December 6, 2001
Work Order No.: NRC-132 Pages 1-98
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
488TH ACRS MEETING
+ + + + +
THURSDAY
DECEMBER 6, 2001
+ + + + +
ROCKVILLE, MARYLAND
+ + + + +
The Advisory Meeting met at the Nuclear
Regulatory Commission, Two White Flint North, Room
2B3, 11545 Rockville Pike, at 8:30 a.m., Dr. George E.
Apostolakis, Chairman, presiding.
PRESENT:
DR. GEORGE E. APOSTOLAKIS, Chairman
DR. MARIO V. BONACA, Vice Chairman
DR. DANA A. POWERS, Member
DR. STEPHEN L. ROSEN, Member
DR. WILLIAM J. SHACK, Member
DR. THOMAS S. KRESS, Member at Large
DR. JOHN D. SIEBER, Member
DR. F. PETER FORD, Member
DR. GRAHAM B. WALLIS, Member. ACRS STAFF PRESENT:
DR. NOEL F. DUDLEY, Executive Director
HOWARD J. LARSON, ACRS
SAM DURAISWAMY, Designated Federal Official
. I-N-D-E-X
AGENDA ITEM PAGE
Opening Remarks by ACRS Chairman . . . . . . . . . 4
NEI 97-06 Steam Generator Program Guidelines . . .10
Proposed Rulemaking for Risk-Informed. . . . . . .75
Revisions to 10 CFR 50.44
Adjourn. . . . . . . . . . . . . . . . . . . . . .98
. P-R-O-C-E-E-D-I-N-G-S
(8:30 a.m.)
VICE CHAIRMAN BONACA: Good morning. The
meeting will now come to order. This is the second
day of the 488th meeting of the Advisory Committee on
Reactor Safeguards.
During today's meeting the Committee will
consider the following: NEI 97-06, Steam Generator
Program Guidelines; Proposed Rulemaking for Risk-
Informed Revisions to 10 CFR 50-44; Standards for
Combustible Gas Control System in Light-Water Cooled
Power Reactors; ACRS/ACNW Office Retreat; 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 written comments from Mr.
Bob Christie of Performance Technology regarding
proposed rule-making for Risk-Informed Revisions to 10
CFR 50.44.
You all have received a copy of Mr.
Christie's statements, and this document would be made
part of the record of this meeting. We have received
no requests for time to make oral statements from
members of the public regarding today's sessions.
A transcript of portions of the meeting is
being kept, and it is requested that the speakers use
one of the microphones, and identify themselves, and
speak with sufficient clarity and volume so that they
can be readily heard.
Before we move to our agenda, I would like
to make an announcement. I am sorry to announce that
Jack Sorensen's term with the ACRS expires on December
31st, 2001. So, Jack will be leaving us. We will
miss him, because Jack over the past four years has
made significant contributions to the ACRS.
He has done work for the ACRS in a number
of areas, including defense in depth, risk informed
regulations, safety culture, and human performance
research.
The latest paper that he developed for us
was regarding risk-informing the GDCs, and we thank
him for all his contributions. Fortunately, Jack is
not going to move far away. He will be working with
the Office of Nuclear Materials Safety and Safeguards,
which is in this building.
So we will have hopefully many
opportunities to meet again with Jack over the next
year or whatever period that you have with that
organization.
I want to especially thank you from George
Apostolakis. He really wanted to be here and
recognize you, but unfortunately had a conflict, and
I am sure that he will be sitting down with you and
thank you personally.
I don't know if you have any statement
that you would like to make.
DR. SORENSEN: Well, thank you for the
acknowledgement. I have thoroughly enjoyed the four
years that I have spent here. I certainly have had
experiences that I would not have had otherwise, like
writing a joint paper with Dana Powers, and Tom Kress,
and George Apostolakis, and a number of other
challenges.
I do thank you for the opportunity to work
with you, and I hope that I will see you periodically
over the next year or so.
(Applause.)
VICE CHAIRMAN BONACA: Okay. With that,
I think we will move to our agenda, and the first item
on the agenda is NEI 97-06, Steam Generator Program
Guidelines. I will now turn to the Chairman of the
Subcommittee, Dr. Peter Ford.
DR. FORD: Thank you, Mario. The last
time the full ACRS subcommittee was briefed on the
industry steam generator program guidelines, and that
is NEI 97-06, was in April of 1999.
Sine then, there has been a hiatus in the
staff's review to the emerging issues associated with
Indian Point, Unit 2, tube failure event, and a
differing professional opinion. The staff's planned
actions in response to the steam generator integrity
issues have been reflected in the Agency's steam
generator action plan, in which we had a part.
In the last six months there have been a
significant number of constructive discussions between
the staff and NEI at all management levels. These
discussions centered around, first, the staff's
comments on EPRI guidelines referenced in NEI 97-06,
steam generator program guidelines; and that was dated
January of 2001.
And, secondly, the NEI generator license
change package, dated February of 2000, which
addresses irregularity aspects of changing the
technical specifications.
There is now substantial agreement between
the staff and NEI. The staff believes that the
revised generic license change package can be
submitted that would offer sufficient regulatory
control.
Some issues have to be resolved. For
instance, management's inspection intervals need to be
justified, and suitable language needs to be included
in the administrative technical specifications in a
change package to provide necessary regulatory
controls on inspection intervals.
However, these issues are not seen as
show-stoppers. Presentations were made to the
Materials and Metallurgy Subcommittee on September
26th, and again on November 29th.
Given that the NEI plans to submit a
revised generic license change package in mid-2002,
and that the staff plans to review and approve the
change package by the end of 2002, we thought it was
timely that a full ACRS committee should be exposed to
the regulatory approach and to the remaining technical
issues, so as to offer advice to the staff and NEI as
they go into the home stretch.
We have specifically asked that the
presentations have data to back up the opinions and
conclusions, and at this point, I will hand it over to
Louise Lund to start off with the staff.
DR. SHACK: Before Louise starts, I would
like to state that I do have a conflict of interest in
this area, because Argonne is doing work on some
generator programs for the NRC.
MS. LUND: Okay. Can everybody hear me?
Okay. Good morning. As Dr. Ford indicated, I am
Louise Lund, and I am the section chief of the
Component Integrity and Chemical Engineering Section.
And what I had hoped to present this
morning is an overview of NEI 97-06, Steam Generator
Program Guidelines, and give you kind of a basis for
which you can understand the next few presentations,
which one is going to be given by EPRI.
And then Jim Riley of NEI is going to make
some comments, and then Emmett Murphy of the NRR staff
is going to also get into more detail in some of the
issues that are in my slides.
Okay. This slide presents a little
history of the staff's activities in the past years in
the steam generator regulatory framework. I think
that most of you are familiar with that, and I am not
going to spend much time on this, because I think that
the committee is aware that there has been a
rulemaking generic letter on different things that we
worked through.
But I wanted to make a quick few points
germane to today's discussion. NEI informed the NRC
in December of 1997 of the industry's intent to commit
to a formal industry initiative called NEI 97-06, with
all PWRs implementing at no later than the first
refueling outage, starting after January 1st, 1999.
What this means is that the industry is
currently implementing NEI 97-06 with the current tech
specs. In the next few slides, I will discuss the
components of the industry initiative of the NEI 97-06
document, the generic license change package, and the
EPRI guidelines, and how these all fit together.
And to then discuss which parts the
industry is currently implementing and what we are
trying to move to. The other thing is that another
reason that I put this together is to kind of give you
a sense for the dates on this, too.
Okay. As promised, here is what the
current framework looks like, and right now as you can
see the current plant tech specs, and the NEI 97-06
steam generator program guidelines that I just
mentioned, what it is, is that it is high level
guidance for utility steam generator programs.
And there is lower tier EPRI guidelines
which give you more of a detailed day-to-day guidance,
and it is referenced in the NEI 97-06 program
document. So the question is what are we moving to.
Okay. This is where we are going, having
the NEI 97-06 program document the EPRI guidelines,
and a generic license change package, which is the new
part of it.
And what it does is that it formalizes the
NEI 97-06 into the NRC regulatory framework via new
tech specs. And how it is envisioned is a generic
license change package will provide a framework for
taking advantage of the flexibility envisioned by NEI
97-06.
As proposed the tech specs and the generic
license change package provide a framework for a fully
performance-based approach. Currently, NEI 97-06 and
the EPRI guidelines are implemented in conjunction
with the existing tech specs which are prescriptive,
with the expectation that soon they will be
implemented with the new tech specs in the new generic
change package.
The advantage to the generic license
change package to industry is a streamline process for
gaining NRC approval of longer steam generator
inspections strategies, alternate tube repair
criteria, and new tube repair methods.
For example, licensees will be able to
implement performance based strategies for determining
inspection intervals which have been reviewed and
approved generically by the staff, without the need
for submitting changes to the tech specs.
The NRC also benefits, in that it is
assured that the steam generator programs will be
focused on tube integrity, rather than simply
following prescriptive surveillance strategies.
And just to kind of give you a sense for
what is in the 97-06 document, the program
incorporates a balance of these items -- prevention,
inspection, evaluation, repair, maintenance, and
leakage monitoring.
And I will discuss also a little bit about
what the generic license change package incorporates,
but it is also important to realize that --
DR. POWERS: Can I ask a question?
MS. LUND: Sure.
DR. POWERS: You say a balance.
Presumably any mix of those things is a balance. Is
there some significance to the word balance?
MS. LUND: Well, I think -- well, you
know, I am not exactly -- I don't think there is
really a terrific amount of significance to the
balance.
DR. POWERS: That's what I thought.
MS. LUND: But I think that there is --
that as you look at it, none of these exist
independent of one another, you know, I guess is
probably a better way to put it.
All of these, as far as your evaluation,
your inspection, and what you repair, all of these
don't exist independent of one another. So you really
need to consider all of them within the same context.
Does that make sense? Do you understand where I am
going? Maybe the wording threw you off.
DR. POWERS: It seems like the most
hopeless -- I mean, there is nothing that I am going
to take home with this. I mean, I am not going to
remember this particular slide for any significant
period of time. Maybe I wasn't intended to.
DR. FORD: Well, maybe a better word might
be interrelationship between these various parts?
They are synergistic to a certain extent.
DR. POWERS: I guess I would understand it
better if I thought I was working with steam generator
tubes that are roughly intact, and don't have lots of
cracks in them. Is this directed towards Alloy 800
tubes, or is this directed towards some better type or
different type of tubes?
MS. LUND: It is intended for all of the
different types.
DR. POWERS: Every one of them?
MS. LUND: Every one of them, and I think
that you need to have a program that makes sure that
not only you consider the inspection and repair, and
leakage monitoring.
It is really a whole package, regardless
of what type of tube material that you are talking
about. It is to make sure that you have a strategy
for dealing with the as-found condition.
DR. POWERS: So if I had a strategy on
repair on detect, I would be okay, right?
MS. LUND: As far as repair on detection
for --
DR. POWERS: Tubes? Let's say I find a
flaw. Fix it.
MS. LUND: Well, that has been one
strategy for doing it in the past.
DR. POWERS: Well, I think the difficulty
with -- I mean, the repair on detection strategy, I
would find that completely unacceptable, because it is
what happens between the time that I fixed everything
and the next time I look that becomes important.
MS. LUND: Yes, which I think is a key to
this and why we are going away from just looking at
surveillance intervals, and going to more of an
approach where we do look at the condition, and we do
look at what has been determined and evaluated as far
as what the operational assessment -- you know, what
the condition of the tubes are and what we predict the
condition of the tubes to be at the end of the cycle.
And so I think that is certainly part of
the new approach, and I think that Emmett is going to
discuss this in more detail when he gets up here. But
that certainly is an important part of it.
DR. POWERS: Okay.
MS. LUND: Okay. And I think that this
one is another thing, and which we will discuss in
more detail later, is that if it does establish
performance criteria that define the basis for steam
generator operability.
And which the performance criteria include
the structural performance criteria, accident leakage
and operational leakage criteria. Now, the generic
license change package, the utilities are to submit
revised tech specs based on NRC approved generic
license change package.
And the generic license change package
will contain a commitment to follow NEI 97-06, and the
licensee's submittal, as you will identify these three
items -- the performance criteria and repair methods,
and inspection interval criteria.
And we will be talking in some detail
later about the inspection interval. How we will do
this is through revised tech spec and bases, and a new
spec with new technical specification steam generator
tube integrity, and a new administrative tech spec,
which is the steam generator program.
This will include revised limiting
conditions for operations specification, and for
operational leakage, and a new limiting condition for
operational steam generator tube integrity.
The new admin tech spec states that the
steam generator program must be implemented to ensure
that tube integrity performance criteria are
maintained.
The licensees will be explicitly required
to assess the conditions of the tubes, versus the
performance criteria, and that shall be performed at
each steam generator inspection outage.
Changes to performance criteria, tube
repair criteria, and repair methods, are subject to
NRC review and approval. Now, just to give you a
little bit of overview as to where we are.
This is the review status, is that the
industry had requested inspection intervals beyond the
current requirements, and where this is going to come
in is actually in Rev. 6 of the APRI steam generator
examination guidelines.
Mohamad Behravesh from EPRI is going to
discuss that in some detail. In the initial part of
the staff review, both the industry and the NRC were
intending a review of performance based inspection
intervals, and as we became aware of changes to the
guidelines, it was apparent that the predictive
methodologies to support performance based inspection
intervals had not been fully developed.
However, the industry is proposing a
reference inspection interval strategy for the newer
steam generator tube materials, and what I am talking
about is 600 thermally treated and Alloy 690 than
those currently allowed in the tech specs.
The industry is currently addressing both
NRC comments and internal industry comments with
respect to this proposal. And the staff believes that
this approach must ensure that tube integrity
performance criteria will continue to be met and that
tubing conditions not meeting the performance criteria
will be promptly detected.
So based on that the staff concluded that
regulatory controls on the inspection intervals were
needed, and what we proposed was incorporating
provisions in the administrative tech spec regarding
the use of NRC approved inspection intervals.
And we have reached agreement with
industry on this approach. So where we are now is
that now that we have reached agreement with industry,
we had a meeting to work out a schedule to reach a
conclusion of the review of the generic license change
package.
So one of the items that we have been
asked to discuss is the schedule and based on recent
meetings, we expect the industry submittal in mid-
2002.
After the submittal is made, we will use
the process of issuing the safety evaluation that we
previously informed the Commission that we would
follow, and that was in SECY-00-0078, and sending it
out for public comment, resolving comments, briefing
the Commission, publishing it in a regulatory issue
summary.
And we anticipate that this will take
approximately six months after we get the industry
submittal.
VICE CHAIRMAN BONACA: Ms. Lund, during
the Subcommittee Meeting that we had last week, we
really had some difficulty with the issue of going to
a performance based, because the ability to determine
tube integrity was not convincing at this stage.
And I think that the NRC is not convinced
either yet. And so we had some concern about this
package coming, and then we had a presentation from
EPRI that said that this is going to be far in the
future.
Right now we are going from a prescriptive
inspection interval process to a better and more
conservative prescriptive process, or something of
that kind.
MS. LUND: Kind of a reference interval is
what I think Emmett is calling it.
VICE CHAIRMAN BONACA: Yes, it was
specific to certain types of steam generators and so
on. And now when I look at your presentation, I get
the message that actually this new package within 2002
will also contain new criteria, new performance based
criteria?
MS. LUND: Well, it is going to include
performance criteria, but the performance criteria,
some of it is just more explicitly called out than
what was called before.
But what I would suggest is that Emmett is
going to actually discuss some of the components of
it, and more of the details of what the performance
criteria are. So that would probably -- and not
wanting to steal his thunder, maybe -- well, would
that be acceptable?
VICE CHAIRMAN BONACA: Sure.
MS. LUND: Okay.
DR. FORD: One of the conclusions from the
meeting last week, Louise, was that if you take away
the conclusion that I came away with, which was that,
yes, in the long term -- and exactly what Mario was
saying, that in the long term, yes, you are heading
for performance based criteria.
But if we don't have the data quality to
get that by the end or the middle of this year, and
that we are going to a prescriptive --
MS. LUND: To an interim stage.
DR. FORD: That has not changed?
MS. LUND: Right. What we are doing is
that as far as inspection intervals, we are going to
have an interim step. Do you want to address that,
Emmett? It looks like --
DR. FORD: Well, so far, you have only
been talking about -- in the words you have been using
or saying performance based.
MS. LUND: That's right, and the
performance based is more on the way that the program
is constructed, and the intent at the beginning was to
make it performance based, including the inspection
intervals.
But the actual issue that we have been
wrestling with has been that we are not ready for the
inspection interval part of it to be performance
based.
DR. FORD: So is it fair to say that the
overall plan that you are proposing is going to be
adaptable enough so that it can take a performance
based, but right now practically it is sticking with
a prescriptive --
MS. LUND: Right, for that issue, because
the program itself based on the condition monitoring
the operational assessment, and looking at the
conditions of your tubes, and looking at how they are
going to perform for the next cycle, and doing all
that evaluation, that is a performance-based part of
the program right now.
And the part that we are still trying to
iron out is this inspection interval strategy. So
does that make sense?
DR. FORD: Yes.
MS. LUND: So I have just one more slide
and then I am going to hand it over for you to get a
lot more details on this. In our long term action for
review status is to resolve outstanding issues with
the EPRI guideline documents, and this will have
future benefits, such as permitting the use of
intended performance based approach for inspection
intervals, in lieu of these reference prescriptive
requirements.
If there is no questions on this overview
part of it, I would like to go ahead and turn it over
for more specifics in the next couple of speakers.
MR. BEHRAVESH: Good morning. I am
Mohamad Behravesh from EPRI, and before I go further,
I would like to get an idea of how much time do I have
for this presentation?
DR. FORD: The whole time that we have is
until 10 o'clock, and Emmett, it depends on how long
you need.
MR. MURPHY: I prepared a presentation
that is very short, and so your questions will drive
the length of it.
DR. FORD: So you are the only two
presenters?
MR. BEHRAVESH: Well, Jim Riley from NEI
will have some concluding remarks.
DR. FORD: How long do you need?
MR. BEHRAVESH: I think it would be a
luxury to have about 25 minutes to 30 minutes. If I
could have that much, that would be good. If I take
less of that, then we will have time for questions.
DR. FORD: Okay. So, 25 minutes with
questions would be perfect.
MR. BEHRAVESH: As Louise mentioned, this
entire framework of NEI 97-06 stands on these various
pillars called guidelines, and the strength of this
framework really depends on the strength of these
guidelines.
And one of these guidelines is the
inspection guideline. A lot of the issues related to
steam generators, at the end it sort of culminates in
inspection questions.
So I would like to spend some time telling
you about this industry document that details
inspection requirements, and where it came from, and
what it all involves.
The steam generator inspection guidelines
goes way back to 20 years or so, and the reason for
it, the impetus for it, was that there really wasn't
any recommendations or any guidance on how to do steam
generator inspections.
And unlike pressure vessels, and unlike
piping, steam ASME was somewhat silent, and to this
date it is still somewhat silent as it relates to
steam generating inspections.
So utilities took the initiative back in
'80, putting a set of requirements together, and over
the years that has continued, and from time to time
they have revised it, and formalized it, and brought
input from vendors to the process.
And they added some very important topics
to it, namely performance demonstration, and it went
on to a prescriptive sampling, and somewhat even
changed the language 2 or 3 years ago that it is
really no longer a recommendation or guidelines, but
rather that it is requirements.
Everything has attached to it the word
"shall" and it is sort of a de facto requirement for
the utilities. So the point that I want to make with
this slide with you is that the set of guidelines that
the industry used for conducting steam generator
inspections is very mature.
It is time tested, and it is field tested,
and it has gone through a whole series of reviews, to
the point that it brings us to Revision 5, and that
has been what utilities have been using for the last
3 or 4 years.
And in the guideline it also says that
every two years it is required that we assess the need
for revision of these guidelines. That is that
Revision 5 of the guidelines, based on utilities input
and other input that we have received, have actually
been pretty good.
It has brought a lot of improvements to
the practice. However, it did lack one thing, and
that is that as utilities went and replaced steam
generators, and went to better materials, this
revision, this addition of the guideline did not make
any distinction between 600 mill annealed material and
600 thermally treated, and 690, the newer material.
And so there was a lot of impetus for
revising this, and for allowing for these new
materials.
DR. POWERS: Is Revision 5 the 1997
version, with all the strong language, the one that
was enforced at Indian Point at the time that they had
their tube rupture?
MR. BEHRAVESH: It was one that they
should have followed when they had the tube rupture.
DR. POWERS: Oh, they should have?
MR. BEHRAVESH: Yes.
DR. POWERS: Okay.
MR. BEHRAVESH: This slide pretty much
says what I mentioned to you, that the guideline
delineates the hows and whats, and all those things as
a result of using an assessment. It is a utility
developed document, and we have started work on
Revision 6 in March of 200, and we expect to finish it
by the middle of next year.
It has lots of things in it, but I am
going to concentrate basically on a couple of issues.
The highlights of Revision 6 is these prescriptive
examinations, and then some issues related to data
quality which came out of the Indian Point 2, as a
result of the experience at Indian Point 2.
We have a new addition of the new draft of
this guideline out for draft out for review, and we
should be receiving --
DR. FORD: Mohamad, could you just expand
a little bit on what you mean by data quality? Is
this laboratory data quality, field data, and what
sort of scatters are we looking at? What are the
quality issues specifically?
MR. BEHRAVESH: Sure.
DR. FORD: If you could go briefly over
that.
MR. BEHRAVESH: Yes. Briefly, it is on-
line monitoring of the data as you are collecting it
to determine whether has the data quality, has the
data deteriorated to the point that it can no longer
be trusted for collecting and making decisions. So
changes have to be made.
And so it is on-line such that you would
know that the data that you have used can be -- that
the data that you have collected can be used, and if
that is not the case, then remedies must be made.
DR. FORD: So it is inspection data that
you are talking about?
MR. BEHRAVESH: Inspection data quality
while you are collecting it.
DR. SHACK: Sort of signal-to-noise?
MR. BEHRAVESH: Signal-to-noise would
certainly be the number one element, yes. Data
quality are other things, too. For example, are you
in the right tube. I don't want to get mundane, but
yes.
DR. WALLIS: Does it say what you do with
the data once you have got it?
MR. BEHRAVESH: Well, what you do with the
data is that you do a whole lot of other things based
on your analysis procedures. But collecting it, and
before you take the instrumentation out and leave the
place, you want to make sure that what you have got is
acceptable and useable.
DR. WALLIS: But you are not addressing
what you do with the data.
MR. BEHRAVESH: Not at this stage, no.
There is a whole set of guidance on what to do with
the data.
DR. FORD: And on your graph beforehand,
or you slide beforehand, you said when. So there is
a time component to this?
MR. BEHRAVESH: Not to the data. A time
component to the data quality, but what I said by when
had to do with prescriptive inspections; when do you
do and how often.
DR. WALLIS: It sounds like a ritual.
DR. POWERS: That is not a bad
characterization.
VICE CHAIRMAN BONACA: Yes, if that is a
condition that is a ritual.
DR. POWERS: You get to anoint a brief to
it, and collect the data, and then they bless it.
VICE CHAIRMAN BONACA: Yes.
DR. WALLIS: Once a day and twice on
Sundays.
DR. POWERS: And every day during a
outage.
VICE CHAIRMAN BONACA: Then you hope and
pray.
DR. POWERS: Twenty-four hours a day,
three shifts.
VICE CHAIRMAN BONACA: And then you hope
and pray.
DR. POWERS: And hope and pray that your
probability of detection is high enough.
VICE CHAIRMAN BONACA: Yes.
MR. BEHRAVESH: Now, let's wait a minute
now. You have to consider the following. It used to
be that you would go there and collect a bunch of
data, and go home, and then worry about what to do
about it.
What is being done now is that know what
it is that you are collecting, and is it what you
intended to collect, and can you use it, and does it
have all the attributes, before you leave the place.
It may sound like a ritual, but it is a good ritual.
DR. WALLIS: Well, did someone go through
the exercise of what you are going to do with the
data, and how much data do you need in order to
satisfy the needs of whatever the process is for
processing the data?
MR. BEHRAVESH: Yes.
DR. WALLIS: And the questions which are
being asked by the whole process?
MR. BEHRAVESH: Yes. There is a whole set
of tables, and the rationale, and tables that says
this is the data that you collect, and this is where
you collect it from, and this is how many times you
collect it, and how much you collect.
DR. WALLIS: And you should deduce that
from your needs for whatever it is or whatever the
questions are that you are asking. The process for
collecting data seems to become lost once you have
figured out what questions you are asking, and how
much data you need, and on what frequency in order to
answer the questions.
MR. BEHRAVESH: That is correct.
DR. WALLIS: Am I right in assuming there
was a process of that type that you went through?
MR. BEHRAVESH: Yes. Now, in a nutshell,
the big changes that are happening in Revision 6 of
the guideline, in comparison to Revision 5, is that
the distinction that has been made in inspection of
600 mill annealed thermally treated, and 690 thermally
treated.
And those distinctions are that for 600
mill annealed material that you inspect it at every
outage. For 600 thermally treated, it must be
inspected at least every other outage.
And for 690 thermally treated, it must be
inspected at least every third outage. And this means
that you cannot skip more than two times, and this
means that you cannot skip more than one outage.
DR. FORD: Last week, we discussed this,
and there are two issues, practical issues.
MR. BEHRAVESH: Yes.
DR. FORD: One is the probability of
detection, and the other one is what is the database
to give you those --
MR. BEHRAVESH: I can get to that and
these are all experience base. This material has
failed in service and is being looked at at every
opportunity.
In this material, there are, oh, 15 or 16
plants in the U.S. that have this material, and they
have gone without any cracking in them for as long as
15 years.
DR. FORD: And did I understand that there
are some incidences of cracking of 600 TT?
MR. BEHRAVESH: Well, two things. There
are alleged cracking in the U.S. that have not been
confirmed. For example, what they thought was
cracking in the Braidwood and Bryon plants that are
600 thermally treated, and they took three tubes out.
And after a destructive assay, they
decided that those were really manufacturing flaws,
and they were not cracks, and they were basically what
is called OD groups. This is in those two plants.
They thought that they had the same or
they thought they had what appeared like cracking at
Turkey Point 3 and 4, and then they went and did an
additional examination with the ultrasonics, and in
reviewing the data and comparing it with the other
similar indications, they again have concluded that
those are not cracks.
Now, there are facts that have been
confirmed in this material in foreign plants, and
there is an explanation for them. For example, in
France, they have quite a few of these, but their
expansion are all very different, done differently,
and what is called the KISS rule, which has introduced
additional residual stresses.
And in those cases, the ones in France,
they have failed in the cracking mode. No such
expansions do exist in the U.S. There may be cases in
Japan where again the thermal treatment has been
different than in the U.S. based on everything that we
know today.
Again, I need to add that this study is
ongoing and we have a study under way to get all of
these details in one place to support these things for
the NRC review.
That in everything that we know today,
this material in the U.S. has not failed in the
cracking mode, and the experience with it in the
field, the number of plants are 13 effective full
power years, and up to 15.
And there are two 30 plants, both of them
are at 15 plus effective power years. So this one is
new and there have been no evidence of failure of it
in the cracking mode in the field. So there is data
being collected for all of this.
DR. POWERS: Your slide implies that the
800 Alloy will be treated the same as the 600
thermally treated?
MR. BEHRAVESH: Pretty much. That is what
we are considering.
DR. POWERS: Is there a reason for
treating 800 the same as 600?
MR. BEHRAVESH: Nothing that I can
elaborate on.
DR. POWERS: I mean, the Germans seem to
be pretty enthusiastic about 800.
MR. BEHRAVESH: Yes, and 600 has not been
bad here either.
DR. POWERS: Yes, apparently not.
DR. SHACK: Do we have any 800 plants in
the U.
MR. BEHRAVESH: Not fully, no. Things are
made with 800 material, but not fully steam generated
plants. Anyway, this is in a nutshell what the
changes are. Now, let's look and see --
DR. FORD: Mohamad, you are going to get
to this question about why these specific multiples in
the inspection periodicities. You are going to get to
it?
MR. BEHRAVESH: Well, I would come to it.
Again, I would come back to it and hopefully be able
to answer your questions. So when we go to 600 mill
annealed, the requirement is that you inspect all of
the tube in every steam generator within 60 effective
full power. And also it means that mechanical --
MR. ROSEN: You are talking about full
power months?
MR. BEHRAVESH: Months, effective full
power months, 60 effective full power months. Steam
generators shall be inspected at every outage, and
every time you do an inspection, you must do at least
20 percent random samples.
If you do samples of less than 20 percent,
this is really no change. This is the way that we
have been doing it in the Revision 5 as well.
DR. WALLIS: Typically, how many months
are there between outages?
MR. BEHRAVESH: It depends on the outages,
which are often times dictated by fuel cycles, and so
you can assume about 12 months, all the way to 24
months, but typically 18 to 22 months or so.
DR. WALLIS: And this is 20 percent
guessed at by some experts or where does it come from?
MR. BEHRAVESH: Well, 20 percent is sort
of the optimum sampling that if you do less, you are
not going to get a lot more information. And if you
do more, it is not going to give you a whole lot more.
So it is backed by -- in fact, it was based on the
study that was done --
DR. WALLIS: So 20 percent if you do it,
say, three times between the time that you inspect
everything, and the chance of missing a cracked tube
is pretty high. Just take the probablistic thing.
You do three measurements of only 20 percent, there is
probably a 50 percent chance --
MR. BEHRAVESH: Well, the idea is that it
doesn't matter how many cracks you have in tubes. You
need to catch one of them to have knowledge of the
tubes being cracked.
DR. WALLIS: But there is quite a chance
that you wouldn't catch a particular cracked tube if
you did that?
MR. BEHRAVESH: A particular cracked tube,
that is correct, but if you find a cracked tube, then
you must expand your inspection all the way to a
hundred percent, which would allow you to catch every
cracked tube.
DR. WALLIS: I am just trying to get the
rationale for it.
DR. POWERS: Graham, do you think that is
true that if I inspect 20 percent randomly selected,
that I have a high probability of having missed a
tube?
DR. WALLIS: Yes.
MR. RILEY: This is Jim Riley from NEI.
Mohamad, would you take a couple of minutes and
explain the degradation assessment? That might help
answer this question I think. The selection of tubes
is not completely random.
MR. BEHRAVESH: Well, as part of this
whole process, you select the tubes to inspect, and
also at the end you must inspect all of them. You
select them based on the knowledge of at which
locations degradation may have the strongest chance of
appearing.
DR. WALLIS: Oh, so it is not random
cracks.
MR. BEHRAVESH: No. Within a given
population, once you have selected a population,
within that population is random. It is random.
VICE CHAIRMAN BONACA: But again going
back to the previous slide, an important bullet is
missing, which is that if you find one, you expand
your inspection beyond the 20 percent?
MR. BEHRAVESH: Oh, yes, exactly that.
VICE CHAIRMAN BONACA: Okay. That's
important.
MR. BEHRAVESH: In 600 thermally treated,
if the steam generators are free from cracking, and
that is a big if, you must inspect all of them first
within 120 effective full power months, and --
DR. WALLIS: If you know that they are
free from cracking, you don't have to do anything.
MR. BEHRAVESH: Pardon me?
DR. WALLIS: It seems to me that if you
know that they are free from cracking, you don't have
to inspect anything.
MR. ROSEN: Really, what is being said
there is that if they have been free from cracking in
the past.
DR. WALLIS: If they were free from
cracking?
MR. ROSEN: Yes.
VICE CHAIRMAN BONACA: If no cracks had
been identified yet.
MR. BEHRAVESH: Once we get back here, if
this material is cracking, you don't go through this.
You don't go through this. The first time around is
120 months, and 120 months is effectively about 10
years.
And what I mentioned to you is that there
are a number of plants that have 15 years of crack
free experience. So that is part of the rationale for
this 120.
Now, the second time around, the
prescriptive methodology is getting more conservative,
and you drop to 90, provided that you live through 220
orderly and completely trouble free.
If you live through that, then the next
period would be 90, and after that, 60 and 60, and 60,
and so forth.
DR. FORD: Mohamad, again looking at this
graph, and the following graph, I come back to my
original question. What is the database upon which
presumably a statistical analysis has been done to
come up with these periodicities? Where is that
database?
MR. BEHRAVESH: That database is the
collection of 15 or 16 plants that have this material
in them, and they have been operating for periods that
exceed this number. So, this number has been taken as
a conservative representation of that experience.
If you have an experience that says 15
years, I would say that at 10 years would be a
conservative representation of that.
DR. FORD: And maybe we are not playing
fair, but have the staff seen that data, and have they
confirmed that those are significant?
MR. BEHRAVESH: Those are public data, and
those are all the --
DR. FORD: Well, has the staff examined
these data? You can have publicly available data, but
have they examined the data?
MR. BEHRAVESH: There has not been a table
that in one page that we have handed it to them, but
they periodically get reports from all of the -- they
continuously get reports from all of these plants that
they know their operational experience.
And the other part of the answer to your
question is that as I mentioned, we are embarking on
a study that is ongoing now to formalize all of this
information; that on one hand appears to be obvious,
and on the other hand, doesn't appear to be obvious to
the uninitiated.
DR. FORD: I guess I am going away with
the impression from this graph and the next one, and
indeed the previous one, that these inspection
periodicities are completely empirical.
MR. BEHRAVESH: Empirical?
DR. FORD: Yes. Is that a fair statement?
MR. BEHRAVESH: Yes, it is, experienced
based. So there is no closed form formula that I can
put some input in it, and out comes on the other side
120.
DR. WALLIS: You could. You could say
that you want to be sure that there is some 99 percent
probability that if there is a crack that you detect
it before it becomes serious, and you determine from
some technical basis how long you can allow it to be
undetected.
DR. FORD: Maybe we can cut this one
short. Emmett, are you going to discuss it?
MR. MURPHY: I don't have viewgraphs that
speak directly to your question.
DR. FORD: But this is one of the concerns
that you have, correct?
MR. MURPHY: Yes.
VICE CHAIRMAN BONACA: Okay. I'm sorry,
but I would like to ask a question of clarification
here. Is the 120 90-60 starting from the
implementation of the tech specs, or from the
construction of the steam generator?
MR. BEHRAVESH: From the replacement or
construction of the steam generator, with a caveat
that after the first cycle of operation -- because
what happens is that once you start, at the end of the
first cycle of operation, you do a hundred percent
examination of everything to make sure that everything
is okay. And from that point onward, this clock
starts.
VICE CHAIRMAN BONACA: Okay. Very good.
Thank you.
DR. FORD: I keep coming back to where
there is data out there, statistical data, and Staehle
has done a lot of this sort of stuff. Does this come
into your argument? I mean, that is scientific data.
MR. BEHRAVESH: Those are all of the ones
that we -- well, we thought that it was obvious, but
apparently they are not. We are collecting it to make
this case as you are mentioning, and that is what I
keep saying, is that it is ongoing now.
VICE CHAIRMAN BONACA: But since you have
no history of cracking whatsoever in the 650, and how
could you possibly have a probablistic basis? I mean,
I don't understand that.
DR. FORD: Some of that is based on the
laboratory data that --
VICE CHAIRMAN BONACA: I'm sorry?
DR. FORD: Some of it is based on the
laboratory data.
VICE CHAIRMAN BONACA: Okay.
MR. ROSEN: Well, just because you have no
failures doesn't mean that you can't do statistics.
You have thousands and thousands of tubes that have
been in service for many years.
VICE CHAIRMAN BONACA: I am talking about
the fact that you are changing from 1210 to 90 to 60,
okay? And you would have to have some criteria of
understanding of how, for example, how cracks initiate
and develop to the point where you can detect them.
I mean, it is not an easy task to simply
understand -- well, all right. Why don't we just hear
from Mohamad further on this.
MR. RILEY: This is Jim Riley again from
NEI. Let me talk a little about where we are going
here. This is a draft document and I would like to
remind everybody of that. The values that have been
put into this document initially are based on
experience.
We have got a bunch of folks who are
working on the documents that are experts in the
field, and they have a good idea of what is going on
out there, both within our country and outside.
And they have put together what they
believe are achievable inspection intervals. The
industry is involved in a study to develop the basis
for these intervals to back up what we are putting in
this document.
Please, right now, these are what we have
in a draft document. We are developing the basis to
prove that they are indeed the right intervals, and we
will have that done before this document is issued in
final form.
There are lots of other things that I
think Mohamad is going to go into that explains not
only the basis, but why these values, even where they
are, are quite conservative.
We have what is called a degradation
assessment, and we have briefly touched on it. A
degradation assessment is required before every steam
generator inspection, and it is required every time
you shut down your plant for a refueling outage, and
you take a look at what is going on in my steam
generators, and even if I am not doing an inspection.
And that degradation assessment looks at
what is going on around the industry, and what is
going on around the world. Do I have anything, any
mechanisms that are going on. If I do, I need to
evaluate how that affects me.
It may change the time that I have until
I do my next steam generator inspection, and it may
change what I look at when I do my inspections. These
degradation assessments plan.
They plan what you are going to look at,
and how you are going to look at it, and what
techniques you are going to use. And they, along as
you are going through your conditions and looking back
and determining what your steam generator looks at,
affects your sampling.
How much do you sample, and what do you
look at, and how do you look at it. All of that then
is rolled up into this operational assessment. The
operational assessment looks at where am I now, and
what is going on, and how long can I go before I have
to worry about exceeding the performance criteria.
That establishes your next inspection
interval. All these values have to be supported by an
operational assessment. So there is a lot of things
that are going on behind the scenes that back up these
values.
DR. FORD: And that's why those things are
synergistic.
MR. RILEY: Yes.
DR. FORD: Sorry, Mohamad, that we got off
on a different subject.
MR. BEHRAVESH: So if they are free from
cracking, and you inspect a hundred percent of them in
these periods -- and again these go way into the
future, and these guidelines are supposed to get
revised every two years, which means that if new
information comes up any time within these, that these
recommendations will change.
DR. WALLIS: I don't understand how that
-- how is that compatible with this bottom bullet
about no SG can operate for more than two refueling
cycles without being inspected. There are lots of
refueling cycles within 120.
VICE CHAIRMAN BONACA: That is confusing,
and I was speaking to that last week, because if it is
a hundred percent of the tubes within 120 months, but
every other cycle you have to inspect some.
DR. SHACK: You inspect a 20 percent
sample.
VICE CHAIRMAN BONACA: So, I guess --
DR. FORD: I guess these are
uncertainties, and I was trying to move it along here,
guys. My feeling is that all of these questions are
tied up with the statistical evidence.
DR. WALLIS: Sure. Of course, they are.
DR. FORD: They haven't shown us that it
will be developed.
DR. WALLIS: Well, I think these numbers
ought to be deduced from something else. Let these
numbers be "X" and you solve from "X" by working out
some other --
DR. POWERS: Well, if it helps at all, if
I assume that the probability in any given tube cracks
is one in a thousand, and I take a sample of a fifth
of the tubes, there is about a 50 percent chance that
I am going to find any cracks.
If it goes up to five out of a thousand,
there is only about a five percent chance that I am
going --
DR. WALLIS: There is one in a thousand
that what you are looking for, or one in 10,000 for
what you are looking for. Are you looking for one
tube or are you looking for 10 tubes typically before
you do anything? It is all tied to your action, and
I want you to find and detect.
MR. SIEBER: Right. Three thousand.
MR. BEHRAVESH: In a steam generator where
you have two or three thousand, or something about
that number of tubes, when you do 20 percent sampling,
if you have 11 tubes in there that are cracked, this
20 percent sampling allows you to catch at least one
of them.
DR. WALLIS: Yes, but is 11 tolerable is
the question. I mean, what is the threshold were you
do something I think is the key question there.
MR. SIEBER: Well, you something when you
get the 40 percent through all, and detect that, and
a single tube in a steam generator isn't going to
corrode and crack all by itself.
The whole steam generator starts to go
back the very day it is born, okay? So when you take
a 20 percent sample, you are going to take a sample of
tubes that are beginning to corrode and crack and so
forth.
And that gives you the confidence to say
I generally know what the condition of the steam
generator is, even though you may not have found a
defective tube, but you will find a corroded tube
after a few years.
And once you find a defective tube, then
all of a sudden --
MR. BEHRAVESH: Well, if expansion comes
into place, then you look at all of them, as opposed
to --
MR. SIEBER: And you look at them and
everything that you can get to.
MR. BEHRAVESH: Yes. The objective is to
detect the onset of degradation. All you need to find
is one degraded tube, and not even defective. One
tube that you believe to be degraded, that puts you in
the expansion of your sampling.
Again, these periods must be supported by
degradation assessment and operational assessment, an
acronym for degradation assessment and operational
assessment, which means that if your degradation
assessment says that you are more likely to have
degradation in your plant, then these things do not
hold. They must be revised.
This is where the industry experience
comes in. You must also satisfy all of the secondary
side requirements, meaning that foreign objects and
things of that sort.
There is also a safeguard in there such
that you don't inspect all of the tubes at the
beginning of the period, and then go on for 10 years
not looking at something, or inspect all of the tubes
at the end of the period, going for 10 years and not
inspecting everything.
And at least 50 percent of them must be
done by about the mid-point within the period, and the
other 50 percent towards the end. And in all of
these, no steam generator can go for more than two
cycles without being looked at.
And then if cracking is discovered in any
time along here, then these things will not hold, and
you go to 600 mill annealed requirements, which means
to look at every steam generator at every refueling
outage.
DR. WALLIS: I could suggest that you skip
this one, and it is exactly the same, but it is just
different numbers.
MR. BEHRAVESH: Different numbers. Now,
the basis for this is that you have 15/16 600
thermally treated, and you have about the same number
of 690 more oncoming on line, and if you look at the
industry as one aggregate of so many steam generators,
you will find out that with more steam generators with
new materials, experience of every one of them must be
considered in the collective experience.
And such that although a steam generator
may not be looked at, although your steam generator
may not be looked at at this time, someone else is
looking at this, and someone else is looking at this,
and all of that information must be considered.
And also if cracking is detected the
inspection interval has to revert back to the 600 mill
annealed, and the second requirement must be met, and
of course this is a considerable enhancement over what
is currently the law of the land, which is the
technical specification, which says that you must do
three percent in about every 40 months.
And it doesn't even say that that has to
be a different three percent. Effectively, you could
go and look at the same three percent sample of tube
within 40 months. So it is a big improvement.
VICE CHAIRMAN BONACA: When you are
looking at this collective experience, I imagine you
are also tracking differences in chemistry, and so on
and so forth?
MR. BEHRAVESH: Yes. That is all part of
the degradation assessment.
DR. WALLIS: Now, the basis for being more
conservative than the current tech spec is presumably
experience, and statistics, or something? Is there
some reason why you have made it more conservative?
MR. BEHRAVESH: Well, if you look at this,
these are plans that have 600 mill annealed material
that have all degraded, and 3 percent would have been
hardly sufficient to detect the onset of that.
DR. WALLIS: So then you can do so kind of
math which says three percent leads ot something
unacceptable?
DR. POWERS: You need another term that we
just don't have I think, Graham, to answer the
question; and that is that given that there is a crack
in a tube that you don't know about, what is the
probability that that tube will rupture during a cycle
of operation. And that is a number that I just don't
know.
MR. BEHRAVESH: That's right.
DR. WALLIS: Yes, but apparently his
experience has led him to change the inspection, and
so there must be some kind of qualitative connection
between what you expect and what you --
MR. BEHRAVESH: Yes. The operational
experience. Remember that I said these guidelines,
and these recommendations have been evolving over a 20
year period.
And so once you started seeing a cracking
through this type of material, it was obvious that the
three percent was not the right number. And I doubt
that it has been exercised by anyone.
On the matter of data quality, is that the
question is the data getting noisy to the point of
effecting detectability and sizing, and this is really
a synopsis of all the objectives for data quality.
And then these examination techniques
specification sheets are all those attributes of a
technique and system and those parameters that you
need to follow.
And then if data gets to get noisy and
affects these parameters, and affects your probability
of detection, and so the question again is to have the
bounds of these examination technique specification
sheets, and if performance has been exceeded such that
the performance indices have been degraded, these are
the basic questions that are being asked.
And as a result, the guidance provides
frequency, location, acceptance criteria, and
corrective action for each of the listed quality
parameters. And these are some of the attributes of
Revision 6 of the guideline.
Currently as I mentioned, too, the draft
of the guideline is outside for review, and we expect
to get comments by December 18th, and start evaluating
those comments early in January, with the hope of
getting this revision out by the middle of the year.
VICE CHAIRMAN BONACA: That is the same
question that we had during the subcommittee meeting,
and again you presented to us a plan that is going
from a description inspection intervals to
prescriptive inspection intervals.
MR. BEHRAVESH: Okay. Let me take a
minute to answer that.
VICE CHAIRMAN BONACA: You didn't say
anything about changing to --
MR. BEHRAVESH: Sure. I was looking at
the time. Let me take a minute to address that.
Everyone talks about prescriptive versus performance
based, and most of them are saying that ultimately we
have to go by way of performance based, where the
prescriptive methodology is not good.
What you need to do in order to do
performance based is that you need two things to do
performance based. You need data and you need
methodology. You can sit around the table and discuss
methodology as long as you want until you agree upon
methodology and put it aside.
When it comes to data, you can't
manufacture data. You can manufacture history, and
you have the data that you have. Now, if you take
that data and put it through the methodology, then are
you willing to accept the outcome.
If the outcome is very good, chances are
that the outcome will be more credible. If the data
is questionable, then the outcome would not be very
good.
The industry recognizes that the NRC
recognizes that the way to go about this performance
based, but right now if you don't have sufficient data
as we have had an example, you put the data that you
have in the methodology, and it comes out and says
here you can go for about 22 years and not look at
this thing.
And everybody says, whoa, wait a minute.
Do I really want to go 22 years without looking at
this, and then you step back, and then the question is
asked, well, what is the problem. Why don't you
accept this.
Is it the methodology that you don't
accept, or is it the data that you don't accept.
Well, data is what you got. There are 690 and there
are two plants in the U.S. that have the experience of
about 8 years or so.
Is that sufficient information, and you
can debate that. Is that good enough information, and
you can go along with it and accept it, and so as
experience, as this collective experience is
accumulated for all these plants, the industry as a
whole will be in a much better position to use that
data and put it in an acceptable methodology to come
up with performance based.
And we have provisions for performance
based within Revision 5 of the guideline, as well as
within Revision 6 of the guideline. In Revision 5, no
one had taken it up.
In Revision 6, some people have sort of
experimented with it, and they are sort of discussing
it, and discussing whether they want to go along with
it or not.
And we anticipate that there will be more
exercises in that arena and with the new agreements
that are being discussed with NRC, there will probably
be some limitations on the maximum interval that the
utility can go without getting that approved by the
NRC.
And I think Jim Riley had some concluding
remarks as part of our collective representation.
MR. ROSEN: Well, I had something to say
on a different issue.
DR. FORD: I will allow 10 minutes at the
end here to ask the Committee here their advice as we
move forward on this. So between the two of you can
you take 10 minutes?
MR. RILEY: I think it is going to take me
longer to walk up there than it is to tell you what I
have to tell you here.
DR. FORD: All right.
MR. RILEY: I am kind of jumping subjects
on you here a little bit. My name is Jim Riley, and
I am NEI's project manager for steam generator issues.
And I came here to talk about another item
that we are addressing as an industry, specifically on
an application of safety factors in one of our
performance criteria.
Now, this is issue has not come up today,
but I will go ahead lay it on you anyway. We are you
know have been working on this generic license change
package for a number of years, and as it is getting
nearer to its completion, we have asked the NSSS
vendors to take a look at what we are putting down
here to make sure that we are consistent with the
design basis requirements.
Let me step back a little and put this
into context. As you have already been told, NEI 97-
06 is the overall document that governs steam
generator programs. Its details are handled through
a whole series of EPRI guidelines.
NEI 97-06 establishes what are called
performance criteria to make sure that your steam
generators continue to operate properly. There is
three of them.
There is a structural integrity
performance criteria which is what its name implies,
and which ensures that the tubes are capable of
withstanding the forces that they are going to need to
withstand in order to prevent their failure.
There is an accident-induced leakage
performance criteria that addresses how much they can
leak, which again is based on design basis
requirements and exposures, and those kinds of things.
And then there is an operational leakage
performance criteria, which is one that you can
actually measure real time while you are operating,
and it is basically just a leakage requirement, a
primary-secondary leakage requirement.
The three of them together act to ensure
or to maximize the probability that you will maintain
tube integrity between steam generator inspections.
And that is the whole goal of what we are doing here.
You know, the old tech specs were very
prescriptive, and very small sample sizes, and they
were based on surveillances that you did every time
you shut down the new steam generator for inspection.
This whole program is set up much more on
maintaining tube integrity. It includes not only a
look back at what my condition is right now, but it
includes methodologies to look forward, and to make
sure that the next time that you shut down and do an
inspection, there is a very high probability that you
will continue to meet your performance criteria.
So within all of this mixture, performance
criteria are very important. They are central. As we
were reviewing the performance criteria this summer,
we found an issue that needs to be addressed, and
right now it is an open issue, and I am not going to
be able to give you much detail at all on the
specifics because we are still looking at it.
But this is a performance criteria for
structural integrity. I won't read the whole thing.
The underlying portion of this is the portion that we
have some questions about, and it has to do with what
safety factor you apply to the loads during an
accident condition.
The current performance criteria says that
you apply a safety factor of 1.4. It turns out that
when we asked the NSSS guys to take a look at it,
there was some questions about how that 1.4 is being
applied under the different designs, and whether that
criteria is accurate for everybody.
Now, all I can tell you at this point is
that we are undertaking some work internal to the
industry to take a look at how this 1.4 is applied,
and may possibly end up revising the performance
criteria to make sure that what it states can be met,
will be met, for all the NSSS designs for the
different steam generator designs that are out there.
We will incorporate that change, that
generic license change package revision as we proceed
for its final submittal to the staff at the middle of
this year.
And we will involve the staff in our
discussions on what we come up with on this
application for safety factors and as we come up with
something more concrete. But this is another open
issue. That's really all that I have to say in the
interest of moving things forward.
DR. FORD: Thank you.
MR. RILEY: If you have any questions on
NEI 97-06, or how this all fits together --
DR. WALLIS: Well, it seems to me that the
testing has got to be related to the performance
criteria.
MR. RILEY: It certainly is.
DR. WALLIS: And somewhere or another
there must be a link that is explicit.
MR. RILEY: The testing that we do is
called condition monitoring, and it is done every time
you do an inspection, and the testing includes all the
stuff that Mohamad was talking about, or that his
guidelines talk about.
DR. WALLIS: He didn't explain how it is
related to the performance criteria.
MR. RILEY: No, there is a different
guideline that is called the integrity assessment
guideline.
DR. WALLIS: But they must be linked. You
don't want to assure the others.
MR. RILEY: Yes. And the results of the
testing gets fed into an integrity assessment that
actually does an analysis of whether or not you meet
the performance criteria now, and whether you will in
the future.
DR. WALLIS: Good. Thank you.
MR. RILEY: You're welcome.
MR. MURPHY: Okay. I am Emmett Murphy of
the Materials and Chemical Engineering Branch, NRR,
and I have a relatively short presentation, but I am
going to go out of sequence right from the get go.
I want to first talk about performance
sine what I say here I think will serve as useful
background purposes for subsequent discussion about
inspection interval issues.
During the briefing that we provided to
the subcommittee last week, questions were raised with
respect to the basis and the adequacy of performance
criteria that are being proposed for inclusion as part
of the generic license change package.
Before answering that question directly,
I would like to just reiterate that it was the intent
-- it was ours and the industry's intent when
developing these performance criteria to maintain
consistency with the original design basis, in terms
of structural margins and in terms of accident induced
leakage.
This is also consistent with the
philosophy of Section XI of the Code, which in the
absence of a specific safety margin criteria, directs
you back to Section III of the Code.
Also, these criteria are consistent with
the current licensing basis of PWRs as embodied in the
specification plugging limits. But are these -- well,
before getting on to your question then, with respect
to the issue that Jim Riley was just discussing
concerning the 1.4., I can't really respond to what he
said yet until we see what they come up with.
All I can tell you right now is that it
has been our intent to maintain consistency with
design basis and with the licensing basis, and to the
extent that we have not fully done that, then of
course we will be open to discussing how we do that,
and I will just leave it there.
With respect to the thrust of your
question from last week, I think we have to turn to
experience, and what does experience tell us about the
adequacy of these criteria.
We have had eight tube ruptures. When you
examine the circumstances of these eight tube
ruptures, the safety margin goal that you are trying
to maintain really was not material to the
circumstances of the rupture.
Five of these ruptures took place because
the utility was not aware of the damage mechanism
taking place in its generators. Irrespective of
structural margins that you are applying to your
analyses, if you are not aware of the mechanism, it is
going to hit you.
On three of the ruptures, the licensee
utility had some knowledge of the damage mechanism,
but he was not adequately managing the program, either
because of inadequate inspections, or inadequate
assessment of how quickly the damage mechanism was
progressing.
Again, a question or an issue of safety
margins was not really material to the circumstances
of the rupture. I think we also need to consider risk
when answering the questions that arose last week.
The staff has examined risk issues
pertaining to tube integrity in NUREG-0844, the steam
generator U.S. side program. Also more recently, in
NUREG-1570. In spite of the eight tube rupture events
to date, the risk associated with spontaneous SG-TR
events has been found to be acceptably low.
Similarly, the risk associated with tube
ruptures which occur as a consequence of transients or
accidents that are within the design basis again have
been found to be acceptably low.
Before going on, and as background for our
next discussion on inspection intervals, I think it is
important to note that this risk experience reflects
the way that steam generators have been managed
through the years.
And one of the important attributes of
that management process has been to inspect steam
generators in general in every refueling outage. Very
few plants have ever had the opportunity to operate
for more than a single fuel cycle between inspections.
When plants were first coming on line in
the late '60s and early '70s, they ran into problems
very quickly, and they didn't have the opportunity to
take advantage of the 40 month provision in the tech
specs.
It has only been in very recent times, and
I don't know how many plants with the latest
replacement generators were able to take advantage of
the 40 months. I think it is only a handful; 1, 2, or
3, something along that order.
And in the main through the years people
have been examining their steam generators every fuel
cycle, and these risk numbers reflect that practice.
And as we contemplate moving to longer intervals, we
have to consider what are the risk implications.
Are we able through inspection and
analysis and tube integrity assessment to ensure that
the necessary margins will be maintained. And a final
point about risk, is that the studies that have been
performed do show that medium performance criteria are
also important from the standpoint of minimizing
severe accident risk beyond design basis type
situations.
However, severe accident risks as you,
know, clearly there is a lot that we don't know about
it as of yet, and it is a subject that is being
assessed as part of the SG action plan.
That program addresses issues addressed or
identified by ACRS in its DPO findings, and it is our
objective that an approved generic license change
package will not increase risk relative to the current
regulatory framework.
We will go now to inspection intervals,
and let me make a couple of observations here up
front. I think as we sat in the back of the room
listening to some of the proceedings it was observed
that perhaps there is some confusion or some confusion
between the performance-based attributes of the
framework, the new framework that we are trying to put
into place in the tech specs, versus a prescriptive
versus performance based strategy for dealing with
inspection intervals.
We are putting into place a framework
which is performed based in the sense that the
specific objective of the program is to ensure that
you are meeting the performance criteria; rather than
to ensure that you are inspecting every so often, and
you are plugging tubes essentially at 40 percent.
But the specific objective is to or the
main objective is to ensure that you are meeting the
performance criteria, and that you periodically assess
the condition of the generators relative to that
performance criteria.
And in that sense we are performance
based. The tech specs do not include the development
of strategies for determining inspection intervals
that are also performance based. That is, you have
determined how long you can operate before you will
exceed the performance criteria.
As we discussed last week, and as we have
been talking about today, so far the guidelines are
not sufficiently well developed to allow people to
make these kinds of determinations accurately.
And so we rely on a prescriptive strategy,
which based on experience expects to prove corrosion
resistance of the material, and give us reasonable
assurance that we will be maintaining the performance
criteria.
Time is very limited, but we have had the
opportunity to look at the strategy, the prescriptive
strategy that Mohamad was presenting earlier, and we
commented on that strategy quite extensively back in
September.
We identified quite a number of issues,
and among them was the importance of receiving as much
information as can be gathered concerning operating
experience with the Alloy 600 thermally treated and
Alloy 690 thermally treated tubing.
We were aware of reported incidences of
cracking abroad, and we didn't know the circumstances,
or even if it was real, or if people were just making
conservative calls. So, we do think that this is an
important consideration as we consider revising the
prescriptive strategy for determining inspection
intervals.
The second important issue has to do with
the definition of active degradation. Mohamad talked
about operating for 2 or 3 cycles, depending upon your
tube material, to the extent that you are free of
active degradation.
Well, that is not quite exactly what they
mean. Active degradation doesn't mean the absence of
degradation activity under the proposal. It does
allow for a certain minimal or a certain level of
degradation.
And we think that there needs to be a
tightening up of that definition to ensure that we
don't have significant degradation actively taking
place, particularly cracking activity, if one is going
to be operating for multiple intervals.
DR. FORD: Just to -- Emmett, because of
time, do I in just reading the rest of that, you
really are reiterating the concerns that we have
independently brought up. Is there anything new with
the "burst" small volumetric flaws? What is that in
relation to, the last bullet?
MR. MURPHY: This is an issue that is not
uniquely related to inspection intervals. It is a
general issue that exists with respect to how you
conduct condition -- as to how you conduct operational
assessment relative to the performance criteria for
small volume flaws.
It is an issue that has existed all along
up to now, and going to a new regulatory framework
doesn't change the significance of this issue, unless
you are going to operate from alternate intervals.
To the extent that the treatment of small
volume flaws relative to performance criteria is to
the extent that that treatment is not appropriate,
then if you go to longer intervals, then you can
aggravate the potential risk associated with that.
So we would need -- that to the extent
that we are going to go along with longer intervals,
we need some interim approach for ensuring adequate
safety margins for small volumetric flaws.
DR. WALLIS: I hate to slow you up --
MR. MURPHY: Well, let me make one more
comment.
DR. WALLIS: This is the first time in
time detectable cracking that I have seen any evidence
that this inspection procedure somehow is related to
what you think of the actual characteristics of the
material.
How does it behave is probably related to
what you do, and this is the first time that I have
seen any kind of mechanical term defined which you are
actually trying to detect.
And so it would be useful if there were
some scientific basis for how you expect things to
perform, and what you do in order to measure various
characteristics of performance.
MR. MURPHY: Well, we have a fleet of
plants out there with advanced materials, and there
will be plants on the leading edge in terms of the
accumulated operating time.
DR. WALLIS: Well, I am not saying that.
I mean, it is not a random process. Either they fail
randomly or there is some kind of physics which says
that they tend to last pretty well up to certain
times, and then they begin to crack.
And then they are all going to crack very
soon after that. That there is some kind of
mechanical basis for how you expect them to perform,
and which has terms associated with it.
And that has got to be related to what you
actually have as a strategy to inspect, and this is
the first time that I have seen a term actually
brought up in the discussion.
MR. MURPHY: Okay. Well, let me just make
one other general point. It is not the objective. It
has never been the staff's objective either with the
old regulatory framework or with this new one to
eliminate the likelihood of future tube ruptures.
The strategies that we have been
considering are not going to accomplish that
objective. It is to maintain the frequency of such
occurrences, tube ruptures or situations where we
don't meet the performance criteria's sufficiency low
frequency, and that the risk implications are
acceptable.
DR. KRESS: Do you have a number for that
sufficiently low frequency that you want to give us?
DR. POWERS: If I back calculate from the
numbers, and assuming that the past frequency is
acceptable, which I think Emmett said, it turns out
that they want to be 99 percent sure that tubes are
not cracking. And it comes out in such a round number
that I can't believe it is an accident.
DR. WALLIS: Well, that must be based on
some model of how they crack.
DR. POWERS: It is strictly a probablistic
model.
DR. WALLIS: Yes, assuming that it is
random.
DR. POWERS: Yes.
DR. WALLIS: But it ain't random.
DR. POWERS: Well, I don't know whether it
is or not, but what struck me as interesting is the --
DR. WALLIS: Well, old tubes are more
likely to crack than new ones presumably. So, it is
not random.
DR. FORD: I'm just forgetting the time.
Mr. Chairman, what leeway do I have on time? I mean,
we can call a halt right now.
VICE CHAIRMAN BONACA: Well, I was just
pointing out that we had a goal to even shorten the
upcoming 50.44, and so your time is very short. I
would say another five minutes, and then that's it.
DR. FORD: So, five minutes?
VICE CHAIRMAN BONACA: Yes.
MR. MURPHY: All right. I would like to
complete a thought. It is not our objective to
prevent tube ruptures entirely. In some small
frequency of tube ruptures and failure to meet the
performance criteria is not going to be unacceptable.
With that in mind, I would argue that it
is possible to entertain the notion of longer
inspection intervals before you developed
methodologies and data, and have data concerning how
long it is going to take to develop degradation
mechanisms that you have not seen before; new cracking
mechanisms and I-690, for example.
How long is it going to take to initiate
those cracks before it may threaten the performance
criteria. So long as the industry works in a
coordinated fashion, and everybody is aware of
everybody else's experience, and to the extent that
one plant is finding cracking, then that information
is disseminated among all the utilities.
And then all the utilities take that
experience into account in determining when they
should be doing their inspections. This kind of
empirical or approach based on experience I think if
properly set up and properly designed will do or will
accomplish the objective of minimizing the frequency
at which you're not meeting the performance criteria
and minimizing risk at acceptable levels.
That concludes my comments.
DR. FORD: Bear in mind that the objective
of this particular presentation was to let the full
committee know where we stood on that particular steam
generator program, and to give advice on any of the
outstanding technical issues that still remain as they
go into the home stretch.
Obviously, we don't have time for us to go
around the table, and I would suggest that we do that
when we come to the letter writing session. Thank
you.
MS. LUND: Excuse me, but could I make one
quick comment? On the handout, I noticed that there
was a typographical error, and I just wanted to bring
that to everybody's attention on the last page.
Emmett, I assume that it is fuel cycles
not to exceed 24 effective full power months, correct?
MR. MURPHY: That's correct.
MS. LUND: Okay. It says EFPY, but it is
actually supposed to be EFPM, and so as long as this
is transcribed, at least we have it in the record,
okay?
VICE CHAIRMAN BONACA: Thank you. Any
other questions or comments from the members? If not,
we will recess now for 15 minutes.
(Whereupon, at 10:01 a.m., the meeting was
recessed, and was resumed at 10:18 a.m.)
CHAIRMAN APOSTOLAKIS: The next topic is
Proposed Rulemaking for Risk Informed Revision to 10
CFR 50.44. Dr. Shack is the Chairman of the
Subcommittee.
DR. SHACK: Okay. We previously reviewed
a staff feasibility study of developing a risk
informed version of 50.44, and we now have a draft
version of the rule, and it is presumably the first
fruits of the Option 3 approach, and Mr. Markley will
be giving us an introduction to the proposed rule.
MR. MARKLEY: Good morning. My name is
Tony Markley, and I work in the Office of Nuclear
Reactor Regulation. I think you know one of my
siblings here at the table.
Before we get too much further, I would
like to introduce to you our project team. We formed
this team back in early March. To my left is Mike
Snodderly, who is in the Division of Systems and
Safety Analysis.
We have in the audience Jim Pulsipher, who
worked with us on the standard review plan and other
input; David Cullison, who worked with the regulatory
guidance; Pete Precinas and Charlie Tinkler from the
Office of Research; Kerri Kavanagh, who worked with us
on tech specs; and a number of other folks who
provided us input on new reactors and so forth.
Basically, we are here today to discuss
the draft proposed rulemaking for Risk Informed 50.44,
and we would appreciate receiving your feedback on our
plans as we present them to you on moving forward with
that rule making.
With respect to the background, as Dr.
Shack indicated, this issue has been before the
Committee in the past in the form of SECY-00-0198,
where a framework for risk informing Part 50
regulations was presented.
In addition, in that paper the
presentation also included some specific
recommendations on how we were going to proceed with
risk informing 50.44, and we also included within that
paper an attachment that addressed some of the issues
that the petitioner, MR. Bob Christie, had raised with
respect to 50.44.
Since that time, we have received a staff
requirements memorandum from the Commission in
January, that directed the staff to proceed
expeditiously with the rulemaking on Risk Informed
50.44.
We sent another paper to the Commission in
August, and that was SECY-01-0162, that specifically
indicated the staff's plans for moving ahead with
50.44.
There were some changes in approach that
were identified in that paper. The first was that as
opposed to providing a voluntary alternative rule
making for 50.44, the staff looked at the fundamental
mission of what RIP50, Option 3, was all about.
That was going back and looking at the
fundamental reason why we had a regulation, and what
does it do, and how does it work. And then once we
had established a need for that regulation, we took
risk insights and risk information to look at the
relative importance of what we think we need for
regulation in that area.
And then to the extent possible
afterwards, we tried to incorporate some performance
based concepts in he regulation, and specify more of
what is to be done, as opposed to how to do it.
So this was discussed in the SECY that
went up in August, and there was another item that
went up and that was a major issue in that paper. One
of the issues that was identified in this RIP50 option
pre-process associated with 50.44 was an issue
associated with the MARK III blowing water reactor
containments, and the PWR ice condensers.
The issue basically involves if you look
at the risk significance sequences for those
facilities that would get you into a severe accident,
if it involves station blackout, then was there a need
for independent or backup power to the igniter systems
of those facilities.
This was identified as a generic safety
issue and has been transmitted to the Office of
Research, and the Office of Research is now in the
process of evaluating that issue.
And that is moving on an independent track
to the rulemaking. So we are not constrained in terms
of moving ahead by the resolution of that generic
safety issue.
With respect to stakeholder interactions,
we have had numerous public meetings on what --
CHAIRMAN APOSTOLAKIS: Are you going to
use your transparencies or --
MR. MARKLEY: I'm sorry. I am going right
ahead here. We will get this up. I had a captive
audience here that was paying attention to my speaking
and no watching the board here.
Now, we did have quite a number of
stakeholder interactions on this subject. We held
meetings and workshops in 1999 and 2000, and based on
the input and feedback that we received back from
that, that has been taken into consideration, in terms
of where we stand on this proposed rule.
We also had a Commission briefing
associated with the Option 3 and some of the other
risk informed initiatives in July. In addition to
that level of stakeholder interaction, the Commission
also asked the staff to make draft rule language
available to stakeholders.
And as of November 14th, we did have draft
rule language for 50.44, as well as some bracketed
information, to try and provide stakeholders some
additional insights into where we were headed with
that regulation, rather than just raw old language on
the paper for them to try and guess at the meaning.
So we received a few comments and
questions and so forth on that, that we have had by
means of telephone communications and so forth, but we
have yet to have any formal comments submitted.
Well, I take that back. We did receive
Mr. Christie's comments yesterday, which I believe you
received as well. So with that, let's go to the next
slide.
Basically where I want to take you with
what we are doing with this rulemaking today is
basically a summary of the changes -- where we are,
and where we are going, and what we believe the rule
is going to look like.
For currently licensed and future reactor
licensees, this rulemaking will eliminate the design
basis accident as a source of significant combustible
gas.
Based on the analysis and the work that
was done in the framework, and previous studies, and
so forth, the design basis LOCA accident just cannot
produce enough combustible gas to challenge
containment.
The second thing that this rule will
accomplish is that it eliminates. Because the design
basis accident in essence goes away, you are more
concerned with beyond design basis, and consider
accidents.
And you no longer have a need for
recombiners, or purge and repressurization systems.
So this rule eliminates the requirements for those.
Once again because the design basis accident is
eliminated, your monitoring systems and other
combustible gas control systems no longer meet the
definition of equipment required to be safety related.
So what this will in essence do will allow
the commercial grade procurement of these items. The
next item. While these are -- well, while the items
that I have discussed already allows a significant
level of burden reduction for current licensees and
future licensees, there are a number of regulations
within 50.44 that are still pertinent. And this
rulemaking endorses those regulations.
DR. SHACK: One thing I noticed is that
most of this is consistent with what you had in the
feasibility study. The one element that I found that
was different was that you were retaining the 75
percent hydrogen generation; whereas, the feasibility
study talked about coming up with a combustible gas
source term that was more mechanistically based.
And you seemed to have punted on that. Is
that a matter of time, or you just decided that it
wasn't worth the effort?
MR. SNODDERLY: Dr. Shack, let me try to
address that. Basically what we found was that the
time frame that had been established for completing
the hydrogen source term was such that it was going to
hold up the rule.
And the Commission challenged us to
present alternatives for speeding up the rule. One of
the options that we presented was to use the 75
percent metal/water reaction, and that has been
produced to the Commission, and we have not heard from
them one way or the other on how we should proceed.
But we were told to continue to proceed,
and we have done so. But we feel that the -- and we
discussed the adequacy of the 75 percent metal/water
reaction in Section 4 of the paper.
And basically what we found was that the
hydrogen source term work that was performed by the
Office of Research -- Charlie Tinkler and Pete
Precinas are here today from the Office of Research to
help support us.
But what we found was that the hydrogen
generation -- that there is two key variables when
designing a hydrogen control system. One is the
amount of hydrogen, and the other is the generation
rate.
And what we found was that the generation
rates that were used in the design of the existing
systems are comparable to the generation rates that
are being calculated in the latest calculations.
So that gives us a level of comfort to go
ahead and continue with the 75 percent metal/water
reaction as an adequate basis for the purposes of this
rulemaking.
MR. MARKLEY: One thing that I would like
to comment on is that the thought and the idea as far
as evaluating the individual source terms, and going
through uncertainty analysis, that effort is still
under way, and is progressing as we speak.
Its function will be more to help resolve
that generic safety issue associated with the MARK
IIIs and PWRI condensers. We will get some additional
information as far as other containment types, but the
initial indications that we have coming back from that
is that even with this source term here, we are still
on the mark for where we need to be with this
regulation.
DR. SHACK: Why embedded in the rule,
rather than -- you know, than putting it in some
guidance document?
MR. MARKLEY: Essentially, this is a
historical construct. One of the things that we are
looking at is that we are risk informing, and we are
trying to make the rule more aligned to the needs that
are out there.
The decision was made to utilize this, as
opposed to trying or as opposed to waiting for perhaps
more detailed information.
DR. SHACK: Well, I can see utilizing it.
But once you put it in a rule though, then it is a
rule change, and you are going to have a Reg guide
associated with this rule anyway. Why not --
MR. MARKLEY: Does it potentially beg
another rulemaking down the road if we get better
information and so forth? Yes, it possibly does.
That is a risk that you do take by putting this in
here.
Okay. The last bullet on this page also
deals with a petition that we received. We became
aware of this petition a little bit later in the
process. Another part of our organization was working
on it, and looking at it.
We had already made a decision to focus
the applicability of this rule on the containment
types, and the challenge to those containments, as
opposed to using specific fuel cladding information.
And whether it is a Zircalloy or ZIRLO, or
that sort of thing. So we had already encompassed
that when we became aware of this. So that is another
change to this rule. And the next slide.
DR. SHACK: And couldn't -- well, couldn't
that potentially affect the source term though?
MR. MARKLEY: We are not so much -- I
think in this case if we are looking at the studies,
the source term that is out there for the large drawn
and sub-atmospherics will not challenge the
containments.
The source term that is utilized in the
MARK I's and MARK II's will not challenge those
containments because those are maintained inerting,
and we are maintaining the requirements for those.
The question is whether the PWR ice
condensers and the MARK IIIs. We don't have a final
answer yet on whether this source term would be
significantly increased for those facilities, or it
could be less for those facilities.
We are relying on this research to be
accomplished, and that research is supporting the
generic safety issue. And that generic safety issue
as I indicated earlier is on its own independent track
for resolution.
And what comes out of that could affect
rulemaking in terms of the PWR ice condensers and BWR
MARK IIIs, or it could affect plant specific backfits
if you will.
You are looking at a very defined set of
population with the BWR MARK III containment types.
I think there is four in the country, and then the ice
condensers, I think there is eight facilities for
those.
So it is a discreet population that you
are dealing with. So our decision at this point is to
let the generic safety issue sort itself out, and what
comes out of that, then we will look at the
implications of that, and whether we need to go to
plant specific backfits, or do we need to get back
into the rulemaking space to adjust what we have here
for the ice condensers and MARK IIIs.
On our next slide, the first bullet for
future reactors is we took advantage of this
opportunity at risk-informing 50.44 to try and
consolidate a lot of the combustible gas regulations.
Part 52 has a reference in there that
endorses the technical requirements of 50-34(f), and
we essentially pulled those at the same level that
they currently exist into the revised 50.44, with two
exceptions.
The exceptions of course being as noted on
the previous page that you don't need recombiners, and
that you are allowed a commercial grade other
combustible gas controlled systems and monitoring
systems.
There were some other conforming changes
that we had to consider when making this change. We
looked at that, and we made a change to -- this says
50.43, and that should be 50.34, and that is a typo on
my part, but 50.34(a)(4).
And we are revising that to make sure that
the need for high-point vents for the reactor coolant
system is addressed in the applicant's PSAR. We are
also revising adding a new paragraph (g) in 50.34 that
will ensure all new applicants include the analysis
required by 54 in their applications.
And the issue with the high-point vents.
This was an interesting issue that we discussed in the
team. It basically came down that the high- point
vents were more associated with proper ECCS
functioning than it really dealt with combustible gas.
Now, it did provide a combustible gas
source term to the containment, which the containment
systems have to deal with, but we felt that this
regulation was really misplaced in 50.44, and so we
are going to move that to 50.46, in a new a paragraph
in 50.46.
Now, in moving this, I will say that there
is an independent effort that is trying to risk-inform
50.46. So we chose not to try and improve the
language or the risk basis for these high-point vents.
Rather, we let the team that was working
on that project have that opportunity. The last item
on this page is the change to 52.47. That is that
reference to the technically pertinent requirements of
50.34(f), the post-TMI stuff.
What we essentially did was say that
except for X, X, and X, that would have referred them
to the requirements if it was combustible gas control
systems.
And essentially that is it with respect to
where we are trying to go with 50.44. Now, we do have
two other slides here that I wanted to present to you,
because we do have two petitions for rulemaking and
you have been apprised certainly of one, if not both
of them.
The first one, of course, is Mr. Christie,
who has made presentations here in the past. The
proposed rule and his petition are significantly
consistent in terms of where he felt the regulation
ought to go, and in terms of what we came up with, and
where we believe the regulation ought to go.
There is only three areas that we differ
a bit, and one is the functional requirement for
hydrogen monitoring. He didn't think that there was
a need for monitoring. Well, from a combustible gas
basis, we would tend to agree with him.
However, the monitoring of combustible gas
in the containment is pertinent in your response to
the emergencies, both for the actions that the
operators would have to take, as well as input into
emergency planning decision makers on whether your
containment barrier is being challenged, and do you
need to initiate protective action recommendations to
either evacuate the surrounding population, or have
them sheltered.
And so the monitors still provide a
significant function, but it just is not combustible
gas control. And since we are looking at a regulatory
efficiency issue, would it be better to just keep all
this monitoring here in one place, or try and put it
in emergency preparedness and so forth, and we look
from an efficiency standpoint to maintain it in the
existing 50.44.
The other area that we differed from Mr.
Christie's petition is the capability for ensuring the
mixed atmosphere. We still feel that is very
important, and that they have either passive or active
means to have a mixed atmosphere, and to preclude
detonations.
We will probably still have the potential
for the flaming or the burning of the hydrogen, but we
would certainly like to preclude the detonation aspect
of it.
DR. SHACK: You had indicated that you
would accept either passive or active means?
MR. MARKLEY: That is the way that it
currently exists.
DR. POWERS: Right. When they propose the
passive means; that is, natural circulation will
accomplish the mixing, what standards of proof do you
require on that?
MR. SNODDERLY: Dr. Powers, I will try and
answer that. I think what we were trying to do here
was to codify the fact that active systems are
important for ensuring a mixed atmosphere.
But during station blackout those systems
are not available, and as part of the IPE process,
licensees were requested to go through their
containments and look for possible vulnerabilities.
So the answer to your question is what we
would like to do is to formalize and put into the
regulatory guide that guidance which is consistent
with that which was used to implement that part of
Generic Letter 88-20.
We want licensees to be aware of places
where possible stratification could occur, and to be
aware of that. And for future designers, we want them
to keep in mind the importance of having an open
containment and want good communication between
compartments to assist in natural circulation.
DR. POWERS: In a station blackout, you
have steam enerting also on your site, and so it is
not so important to have a well-mixed atmosphere
there; is that true?
MR. SNODDERLY: Yes. But again I think
the purpose of the rule is to communicate the fact
that mixing is an important aspect of reducing the
risk from combustible gas, and you can accomplish that
by active systems, such as fan coolers and sprays.
But in the absence of those systems, we
also need to look at the fact that you may be
susceptible because of stratification. Now, the fact
that you have steam is also going to be on your side,
per se, but as time goes on that steam will be
condensing out.
And certainly if you engage containment
sprays you are going to be taking the steam out of it
to some extent, and you will have some separation of
the hydrogen. I think that is one of the insights
that we have also gained from the Office of Research.
And that is that if you were in station
blackout situation, and then if you did get power back
and you decided to use the sprays, it is important --
well, one aspect would be to bring one train on at a
time, as opposed to two trains.
There was some experiments done at Sandia
that looked at the fact that if you got sprays back
and you quickly took out the steam, could you go into
the detonable region.
And what they found was that for
prototypical spray rates that it did not appear to be
a vulnerability.
DR. POWERS: I wouldn't think so. I mean,
mixing -- I mean, you turn on the sprays, and you
condense the steam, but in doing that you do an awful
lot of mixing.
MR. SNODDERLY: That's right.
DR. POWERS: And so I guess what I am
driving at really is that I am trying to see how this
plays out looking at like an AP600 type design, where
you have a feed up until a cooled dome, and you are
relying on the flow in that dome region to keep from
ending up with a stratified hydrogen layer at the top.
MR. SNODDERLY: Yes. We wrote the rule in
a way that the licensee could meet it with that type
of an approach, and we would expect the analyses that
were done in support of AP600 to meet that
requirement.
Now, for current plants though, we would
not expect them to do anything more than what they
have done --
DR. POWERS: Well, you have really
answered my question, which was were you thinking
about AP1000.
MR. SNODDERLY: Yes.
DR. POWERS: And the answer is yes.
MR. SNODDERLY: Yes.
DR. POWERS: Okay.
MR. ROSEN: Your presentation is silent on
the impact of these activities on technical
specifications. Could you tell me what you think
about that?
MR. MARKLEY: I think actually that there
is going to be a lot of benefit to technical
specifications. We included in the package some draft
changes, and essentially all the text specs with
respect to hydrogen monitors are being proposed for
elimination because this is more in terms of the
emergency response and that type of activity.
And we don't normally have emergency
response equipment in tech specs. With respect to the
oxygen monitors for the inerted containments, we are
proposing to maintain those in technical
specifications, because how do you know if your
containment is inerted unless you have a monitor that
would provide that indication.
And so we would look for that to still be
retained in tech specs.
MR. SNODDERLY: And if I could add that we
have Kerri Kavanagh here from the Standard Technical
Specifications Branch, and she has been a member of
the team and has helped us in this area.
What is interesting is that if you look at
5-36 and the four criteria for tech specs, the fourth
criteria talks about demonstrating that the structure
system or component is needed in tech specs because of
risk significance.
This is a criteria that is not used very
much at all, but we used it in this case in trying to
make the determination of what stays in tech specs and
what doesn't.
As Tony mentioned, we have determined that
an inerted environment for MARK I's and II's is risk
significant. That would stay in the tech specs, and
the oxygen monitors for verifying that would also
stay.
But now the basis will change slightly,
and instead of being there for design basis accidents,
it will be there for Criteria 4, and the fact that it
shouldn't be risk significant.
Now, if you look in the past, hydrogen
monitoring was there for design basis accidents, and
to know when to turn on the recombiners, or when you
might have to use the vent and purge system.
Now what we have found is that -- and
maybe also what was needed was to actuate hydrogen
igniter systems in MARK IIIs and ice condensers.
Well, there is other indications that you can use to
determine when you should turn those on.
And as Tony said, the hydrogen monitors
now are more for core damage assessment and emergency
planning, and to be consistent, they would not lead to
-- hydrogen has not been shown -- hydrogen combustion
has not been shown to be a risk significant threat to
containments because of the mitigative features that
have been put in.
So that made us say, okay, we don't need
hydrogen monitors to actuate those mitigative
features, but we do feel that we need them to support
adequate severe accident management, and therefore, we
felt that it was supported removing these from the
technical specifications.
MR. ROSEN: Well, your comments have been
responsive with regard to monitoring, but what about
control?
MR. SNODDERLY: Control? No changes.
Igniters will stay in the tech specs, and inerting
from MARK I's and II's will stay in the tech specs
because those systems have been shown to be risk
significant in the absence of those systems. So
that's why they need to stay in the specifications.
MR. ROSEN: What about -- well, control in
large dry containments, PWR?
MR. SNODDERLY: There aren't -- currently
the only -- the recombiners are in the tech specs, and
we proposed to remove those from the tech specs and
large dry containments because they would not impact
risk significant source terms.
The problem that you have is that the
recombiners work at 200 standard cubic feet per
minute, and when you talk about a large dry
containment with a volume of 2 million cubic feet, it
takes seven days to turn over that containment volume.
And so the recombiners were not shown to mitigate the
consequences of --
MR. MARKLEY: Basically, they are very
ineffective.
MR. ROSEN: So for summary in a large dry
containment, you would not be required to continue to
operate the recombiners.
MR. MARKLEY: That's correct.
MR. ROSEN: And they could be moved from
the plant effectively?
MR. MARKLEY: That's correct.
MR. ROSEN: And the control or the
monitoring would be required to remain, but would not
be safety grade anymore?
MR. MARKLEY: That's correct.
MR. SNODDERLY: Exactly. It would just be
a functional requirement and it would be removed from
the tech specs and it could be met with commercial
grade.
MR. ROSEN: Thank you.
MR. MARKLEY: The last item on this page
is that we did explain the basis for the exceptions
that we took in the statement of considerations. The
last slide that I have is with respect to the NEI
petition.
And as we had stated previously, they were
concerned about the restrictive applicability language
of using Zircalloy and ZIRLO, claddingness, and the
applicability entry for the existing 50-44.
And as I indicated earlier, we early on
decided that that was not an appropriate applicability
statement for what we were trying to deal with in
terms of the subject matter.
So as you have seen in our discussions
today, the new rule as we propose it focuses on
containment types for the applicability of the various
technical requirements.
There is going to be many different types
of fuel come up, and certainly the pebble bed reactor
and other things of that nature, and there is other
things that we probably can't perceive at this time.
So this concludes our presentation, and I
would certainly be happy to entertain any additional
questions.
DR. SHACK: Any additional questions from
any of the Committee Members? There being none, then
thank you for a very clear and lucid presentation on
the proposed rule.
MR. MARKLEY: Thank you.
CHAIRMAN APOSTOLAKIS: Okay. So we will
recess until 1:30.
(Whereupon, at 10:50 a.m., the meeting was
recessed.)
Page Last Reviewed/Updated Monday, August 15, 2016