ML20151J481
| ML20151J481 | |
| Person / Time | |
|---|---|
| Issue date: | 07/30/1997 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-3006, NUDOCS 9708050160 | |
| Download: ML20151J481 (65) | |
Text
RCHS7~3co(,
Official Transcript sf Proemdings O
NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Thermal Hydraulic Phenomena Subcommittee lt TRO4 (ACRS)
RETURN ORIGINAL Docket Number:
(not applicable)
To sawarTE M/S T-2E26 415-7130 THANKS!
l Location:
Rockville, Maryland O
Date:
Wednesday, July 30,1997 lI l
g g j f g,_,_ g g } / g 6 g$gjLeg h em M gl Pages 435-486 Work Order No.:
NRC-1188 Cloud. 19o-4'34 NEAL R. GROSS AND CO., INC.
l*
Court Reporters and Transcribers e
1323 Rhode Island Avenue, N.W.
i b g t c. n p'
Washington, D.C. 20005 fjO l0
"i t
i (202) 234-4433 jf j j ']
jm 9708050160 970730 llllllllllllllll llllllllll{l l[lll ACRS gD 4
.o
bC Nh60%
Official Transcriptrf Proceedings O
NUCLEAR REGULATORY COMMISSION 1
Title:
Advisory Committee on Reactor Safeguards Thermal Hydraulic Phenomena Subcommittee y
TRO4 (ACRS)
RETURN ORIGINAL Docket Number:
(not applicable)
To sawaITE M/S T-2E26 415-7130 TIDJ:KS !
Location:
Rockville, Maryland o
O Dat' :
Wednesday, July 30,1997.
e 1-
)
h' T T]:
[
Work Order No.:
NRC-1188 M gl Pages 435-486 clo.sede Mo-484 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers A
^ t-1323 Rhode Islznd Avenue, N.W.
~,
(.._)
Washington, D.C. 20005
..,jO k. 9!
U (202) 234-4433 y!
!]
[Jg llllllllllll!l llllll!lllllll llllll so O 970730 T-3006 PDR
.._.__m
- O DISCLAIMER' PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSION'S ADVISORY COMMITTEE ON REACTOR SAFEGUARDS JULY 30, 1997 i
The contents of this transcript of the l
proceedings of the United States Nuclear Regulatory l
O Commission's Advisory Committee on Reactor Safeguards on JULY l
30, 1997, as reported herein, is a record of the discussions j
l recorded at the meeting held on the above'date.
j This transcript has not been reviewed, corrected I
and edited and it may contain inaccuracies.
l I
3
.O r
l^
COURT REPORTERS ANDTRANSCRIBERS NEAL R. GROSS j
1323 RilODEISLAND AVENUE.NW WASII!NOTON,D.C. 20005 (202)234-4433 (202)234-443 4
435 1
UNITED STATES OF AMERICA 2
+ + +++
(x EI 3
NUCLEAR REGULATORY COMMISSION 4
+ + +++
5 MEETING 6
+++ + +
7 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
THERMAL HYDRAULIC PHENOMENON SUBCOMMITTEE 9
+ + +++
10 OPEN SESSION 11
+ + +++
12 WEDNESDAY, 13 JULY 30, 1997
v) 14
+ + +++
15 ROCKVILLE, MARYLAND l
16
+ + +++
l 17 The subcommittee met at the Nuclear Regulatory 18 Commission, Two White Flint North, Room T2B3, 11545 19 Rockville Pike, at 8:30 a.m.,
Thomas S.
Kress, Chairman, 20 presiding.
l l
21 l
22 PRESENT:
23 COMMITTEE MEMBERS 24 THOMAS S.
KRESS, Chairman
)
25 MARIO FONTANA, Member NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
436 1
ACRS STAFF 2
SAM DURAISWAMY
< ~g
?
)
3 RICHARD P.
SAVIO 4
PAUL BOEHNERT 5
MEDHAT M.
El-ZEFTAWY 6
AMARJIT SINGH 7
8 ACRS CONSULTANTS PRESENT:
9 10 IVAN CATTON 11 VIRGIL SCHROCK 12 13 ALSO PRESENT:
(%
'w '
14 Ralph Landry 15 Bill Huffman 16 Earl Novendstern 17 Bob Kemper 18 Brian McInyre 19 Mike Young 20 Andy Gagnon
." 1 Jim Lyons 22 Ted Quay 23 Alan Levin 24
(~)
'%y 25 l
l NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
437 1
A-G-E-N-D-A i
2 j p AGENDA ITEM PAGE LJ 3
NRR Presentation 4
R.
Landry 437 5
Subcommittee Caucus 467 6
7 8
l 9
10 11 12 i
l 13 14 l
15 1
r 16 17 18 19 20 21 i
l 22 23 24
,O Q) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 l
l
438 1
P-R-O-C-E-E-D-I-N-G-S 2
g-]
(1:31 p.m.)
3 MR. LANDRY:
Okay.
Again, I'm Ralph Landry 4
from the Reactor Systems Branch.
And what I would like to 5
do is summarize where the staff is in the review of the 6
NOTRUMP code and respond to some of the questions which 7
the consultants asked yesterday.
8 (Slide) 9 MR. LANDRY:
The package of literature which 10 I've handed out has a lot more material than I had really 1
11 decided to go through this afternoon.
I'd like to skip 1
12 over some of it very quickly; in particular, the list of i
13 the topics that I'm going to cover, and just very briefly
,t,i
'!'/
14 talk about the background, which we talked about yesterday 15 quite a bit.
l 16 (Slide) 17 MR. LANDRY:
The basis for the review of the 18 code, again, was the NOTRUMP code applicability document, 19 which we received in 1994.
We said that the CAD was an l
l 20 inadequate description of the code modification, the PIRT l
21 lacked sufficient detail, the code hardware models only i
22 received a cursory discussion, and code phenomenological 23 model changes were not even discussed.
24 We ended up asking an extensive number of
(),)
f 25 requests for additional information.
And in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
439 1
supplemental draft SER, which was written a year ago, we eN 2
had a new PIRT.
We also had commitments from Westinghouse
(
)
%J 3
to assess phenomenological models, perform benchmark 4
calculations, and recalculate a number of the integral 5
systems tests.
6 Very briefly, on the PIRT, the original PIRT 7
we said was unacceptable.
Westinghouse revised the PIRT 8
and in the revised PIRT had four intervals:
- blowdown, 9
natural circulation, ADS blowdown, and IRWST injection 10 cooling.
11 The NRC staff had a PIRT which contained five i
12 intervals.
Two of the intervals in the NRC staff PIRT 13 were meshed into one interval within the Westingh-use p_
(
)
N/
14 PIRT.
15 All high-ranked items were ranked hic. in both 16 of them.
All of the medium items which the NRC staff had 17 in their PIRT for either medium or low, those two or three 18 that were ranked low by Westinghouse were actually covered 19 by the test program and received analysis through the 20 NOTRUMP assessment program anyway.
So they were covered.
21 As long as they were covered, we were satisfied that all 22 of the high and all of the medium-ranked items were 23 covered in the assessment and were captured in the 24 assessment program.
,Oi
(,)
25 So the staff concluded that the PIRT as NEAL R. GROSS f
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE. N.W (202) 234-4433 WASHINGTON, D C. 20005-1701 (202) 234-4433 l
440 1
presented was acceptable or was applicable to the AP600.
2 (Slide) f>',- m
\\
i 3
MR. LANDRY:
Some of the component models 4
which received copy review.
The ADS, NOTRUMP tended to 5
overpredict the flow rates in OSU and some of the SPES-2 6
tests.
At that time, our understanding was that a 7
possible source for the overprediction was related to the 8
characterization of flow regimes in the piping.
We were 9
waiting for the final V and V.
And we are still reviewing 10 that material.
So we don't have a conclusion at t rtis 11 point.
12 The CMT is ranked high in the flow regime in 13 the PIRT.
A big concern that we have and which the
,e O
14 Committee has also expressed is the lack of thermal 15 stratification.
The applicant has gone through a number 16 of methods to deal with the lack of thermal 17 stratification, such as extensive nodalization of the CMT.
18 We note that NOTRUMP does tend to predict a 19 reasonable discharge flow.
However, we're concerned about 20 the lack of the stratification at this point, and we're 21 still reviewing material in the final V and V report.
22 PRHR-IRWST combination, the heat transfer 23 correlations have been modified to include the Shah 24 condensation correlation and the Lienhard and Dhir
,,j 25 modification of the Zuber correlation.
We are reviewing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 l
\\
441 1
the final V and V for the acceptability of the 2
modifications that they have made.
,s l
\\
3 Reactor coolant pumps we said had to be 4
assessed against benchmark calculations.
They had to be 5
benchmark because these are a different type of coolant 6
pump than is in the current generation of Westinghouse 7
PWR.
8 We have noted that the large number of ADS and 9
CMT component tests were committed to for reanalysis.
We 10 are currently reviewing those reanalyses and the 11 information which has been provided in the final V and V 12 report.
13 (Slide)
's /
14 MR. LANDRY:
Some of the phenomenological 15 models.
As I discussed yesterday, we learned in reviewing 16 the test analysis reports that there were 18-some 17 phenomenological models in the code that were enhanced, 18 modified, or altered.
19 We also note that out of those, three of those 20 modifications were changed.
And there were some other 21 modifications that were made as a result of the assessment 22 program, which has been going on on the code.
23 The assessment of models was required by the 24 staff, resulted in the further changes in those models.
!)
25 (Slide) x.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
442 1
MR. LANDRY:
This is just a quick listing.
l l
2 You can get this listing just by looking in the table of fs i
4 s
3 conterts for Volume I of the final V and V report.
It 4
includes drift flux, which Dr. Catton has discussed quite 5
a bit in here.
6 Some of the other items, the horizontal flow 7
mumble mumble model.
Is that the proper term?
8 DR. CATTON:
That's right.
9 MR. LANDRY:
The choking models.
A number of 10 these models, Henry-Fauske and critical flow model, these 11 are the models that have been changed into code and have 12 been subjected to additional assessment.
13 (Slide)
/
\\
\\_
'l L
14 MR. LANDRY:
One of the points that was 15 brought up yesterday, what makes an Appendix K model?
16 Well, to try to cover this rather briefly, Appendix K does 17 not stipulate large break cr small break.
There are no 18 words that say this is a large break, this is pertinent to 19 a small break.
20 Appendix K is nothing more than one way in 21 which a licensee or an application can satisfy the 22 requirements of 10 CFR 50.46.
10 CFR 50.46, of course, 23 says that you have to perform an analysis of a l
24 loss-of-coolant accident.
p)
(_,
25 You must use an acceptable model, which can i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. O C. 20005-3701 (202) 234-4433
443 1
either be an Appendix K or a best estimate statistical s
2 approach.
And you must cover a range of break sizes to f
r b
3 demonstrate that you have captured the worst case or the 4
limiting case as far as peak clad temperature is concerned 5
and core damage if core damage does occur and then 6
stipulates what is the upper limit for a peak clad 7
temperature, upper limit for degradation of the core.
8 Following the Three Mile Island event, the 9
lessons learned from that event resulted in the staff 10 issuing NUREG-0611, which pertains specifically to 11 Westinghouse PWRs.
0611 points out that the Westinghouse 12 flash models were considered deficient for a small-break 13 LOCAs.
p\\
\\
]
'N 14 Shortly after that, staff issued NUREG-0737, 15 which further clarified the small-break LOCA requirements.
16 And, in particular,Section II.K.3.30 required compliance 17 with the guidelines issued in NUREG-0611.
18 Now, there's a further section, II.K.3.31, 19 which says that all licensees and applicants must either 20 submit a small-break LOCA analysis that is demonstrated to 21 be in compliance with the requirements of II.K.3.30 and 22 compliance with NUREG-0611 or they must justify why they l
23 are not submitting a new analysis.
24 A few of the requirements of Section
,9 i
s
(,/
25 II.K.3.30.
There are about 10-12 items listed in l
NEAL R. GROSS I
COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 i
I I
444 1
II.K.3.30.
I don't want to go through all of those.
I'm j
,x S 2
going to touch on a couple of them that pertain to the l
\\,_/
3 3
discussion the last two days.
4 DR. CATTON:
So what's happened in the interim j
5 period?
This second one under that bottom, " Validate 6
condensation heat transfer and effects of non-condensible 7
gases," I don't believe that's been done yet.
8 MR. LANDRY:
I'm going to get to that item.
9 DR. CATTON:
Okay.
10 MR. LANDRY:
You're taking things out of 11 order.
12 DR. CATTON:
Okay.
13 MR. LANDRY:
The first one, " Provide
\\
'/
14 calculated validation of core heat transfer and two-phase 15 level," specifically, as far as the two-phase level, this 16 is what we're calling two-phase level swell.
17 (Slide) 18 MR. LANDRY:
The staff insisted that 19 Westinghouse demonstrate the validity of NOTRUMP's 20 capability of calculating two-phase level swell under the 21 low-flow, low-pressure conditions that would be typical of 22 the AP600 design.
23 Now, these two-phase level swell calculations 24 had been validated for the operating plants because all f3
(_)
25 the calculations were done in the mid to upper-pressure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
445 1
range.
Here we have to assess low-pressure, low-flow s
2 conditions and after much negotiation back and forth with
)
(/
3 Westinghouse arrived at a set of tests that we felt could 4
be useful to assess the capability of the code in this 5
area.
6 We felt that, as we discussed yesterday, one i
7 test program was not sufficient for a number of reasons.
8 The G-2 tests were run from moderate to low power, 9
moderate to low pressure.
However, they were a constant 10 pressure boiloff test, constant break size boiloff test.
11 The GE tests were variable break size tests, 12 but they were a simple vessel blowdown test.
And they 13 were all run at uloderate pressure, 1,000 psi.
7__
t I
U 14 We also looked at a couple of other systems 15 and finally settled on the Achilles test because they were 16 an integral system test, had run at low pressure, and a 17 design typical of the Westinghouse PWR.
18 (Slide) 1
(
19 MR. LANDRY:
From the G-2 test series, we t
20 agreed on two tests each, two different powers, at 21 approximately 780 psi, 400 psi, 100 psi, 50 psi, and 15 l
l l
22 psia.
That gave us the coverage from the moderate range, 23 from just below 1,000 psi down to one atmosphere of 24 pressure.
p
()
25 The GE test, we selected two break sizes at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C 20005-3701 (202) 234-4433
446 1
1,000 psi to be run with the flow restrictor in place and 2
representing the core restriction and without that
-,x I
)
~
3 restrictor in place.
4 We also agreed on two of the Achilles tests, 5
at 30 psia and 18, approximately 2 atmospheres and one 6
atmosphere.
7 Between these tests, we believed that we would 8
have adequate coverage of the phenomena that could occur 9
and could influence two-phase level swell in the AP600 10 design.
Taken individually, these tests had problems.
11 The G-2 test was a boildown of a 17 by 17 12 bundle, but that bundle was within a baffle plate.
The 13 baffle plate had leakage.
The leakage was not measured.
p)
'w /
14 The leakage would flow out the baffle plate, recirculate, i
15 and come back in the bottom of the core.
When 16 Westinghouse did the calculations, they had to do 17 calculations assuming a high and a low leakage and see how 18 those calculations compared with the data.
19 The GE test, as you heard in the discussion 20 yesterday, apparently had a high entrainment of liquid at 21 the break, which was not supposed to occur.
The 22 assumption was that it was dry steam blowing out the 23 break.
l j
24 The only way that the two calculations could
()
2E be matched was to artificially remove liquid mass from the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
447 1
bottom of the vessel, get the mass right, and the level 2
swell appeared to be pretty reasonable.
,3 i
t 3
This has some justification in that the test 4
analyses which were done using the TRAC code shortly after 5
the tests were performed did virtually the same procedure, 6
where Westinghouse found one mass.
7 I believe it was on the order of 20 pounds of 8
mass that had to be removed through an artificial removal 9
system.
The TRAC code was estimating 17 or 18 pounds of 10 mass that had to be removed, very close, 2 different 11 codes, similar procedure, remove about the same amount of l
12 mass.
You get a reasonable calculation.
But-this is a 13 problem if you take the test program in itself, ignore all i (,-,)
l V' 14 other tests.
15 And the problem with the Achilles test program i
16 was the program was not intended to be a two-phase level 17 swell test.
There was a boildown test in a system similar 18 to the Sizewell B plant.
19 However, they did not measure the level of the l
20 liquid in the vessel.
So that you did not know where the 21 liquid level was until you hit a thermocouple at the top l
22 of the core plate.
When the liquid level hit the 23 thermocouple, the temperature would go up.
You knew where 24 the level was at this point.
',q
\\s_,)
25 So the calculation of the test series had to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
448 1
be performed on the fly.
Wait until you come down in time 73 2
conditions until you see where that thermocouple dryout 3
occurs.
Now we're going to initiate the NOTRUMP 4
calculation, start the calculation.
This is what we're 5
calling time zero.
So it's giving an artificial 6
discarding point to the problem.
7 Each of these test programs had a problem with 8
it.
The test programs were not intended to be 9
low-pressure, low-flow rate, two-phase level swell test 10 programs.
But these were the best data that we thought we 11 could find available at this time to do the assessment of i
12 the code.
We thought that, taken together as a group, 13 they would give us a confidence in the code's ability to
(
j
\\'
14 calculate two-phase level swell in the core.
15 Now, on the other side of the question, the 16 AP600 design is not predicted to uncover the core.
So do 17 we need this?
Well, we feel we do because if you did
)
18 uncover the core, could you predict it?
Well, without j
19 assessing these models, you really don't have a clue as to 20 what's going on.
So we insisted on the assessment, even 21 though the core is not predicted to uncover.
22 (Slide) 23 MR. LANDRY:
Now we'll talk about
.2.
In l
24 particular, let me address the non-condensible gases since
,O
(_,)
25 Dr. Schrock was raising that question yesterday.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
449 1
One of the concerns that we've expressed -- we ex 2
expressed it after the CAD and after the test programs --
3 was the lack of a non-condenstb]e gas model in the NOTRUMP 4
code.
We raised this question through RAIs and through a 5
number of discussions with Westinghouse.
6 The argument, the discussion has been that in 7
the test programs, the injection of non-condensible gases 8
does not occur until late in the transient.
The transient 9
is essentially over.
So the non-condensibles become 10 unimportant.
11 Well, the test programs don't have the ability 12 to track non-condensible gases anyway.
So what good would 13 it do to know where they're going?
For code assessment, i
\\
\\-
14 you don't know where they're going.
So you can't tell how 15 good your prediction is.
16 But if we get to the end of the transient, we 17 have found that there were non-condensible gases in the 18 top of the PRHR.
Some went out the break legs of the 19 system.
20 So we're not pleased about this.
However, the 21 code was approved in the past by the staff for use for 22 small-break LOCA, in spite of the fact that it does not 23 have a non-condensible gas model in the code.
24 So the only answer I can give is it appearn
'm
(,,-)
25 from looking at the assessment program that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
p l
450 l
1 non-condensibles do not have a de' gradation effect on the i
2 response of the system.
Even though the model does not 7.- 3
)
l 3
have that capability to calculate and track 4
non-condensible gases, we don't believe that it will be a 5
problem at this point in time.
6 DR. SCHROCK:
The validation report that 7
Westinghouse provides says that it's not a problem because d
there are multiple paths the non-condensible can take to 9
escape from the syetem.
10 But when thcv were asked to explain that, they 11 did not provide an explanation that was satisfactory in my 12 mind.
Is it satisfactory in your mind?
13 MR. LANDRY:
Well, it wasn't a very good
/^N s_-
14 explanation, no.
The c.
- s as we have been able to figure 15 out would distribute throughout in a number of locations 16 in the system and wou:d not seriously degrade capability 17 to recover f om a small-break LOCA.
18
':le have to keep in perspective that the core 19 is not uncovering in these transients and that the core is 20 not heatir up in these transients.
The core does not e
21 heat up appreciably beyond normal operating temperature.
22 DR. CCHROCK:
I'm not making a judgment that 23 the non-condens oles are a problem.
What I'm pointing out 24 is that the document that they are not a problem is
(~)%
25 nonsense.
And so I don't think you're in a firm position
(_
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2 % 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
451 1
to say that it is a non-problem.
2 MR. LANDRY:
I think Westinghouse has heard 73 i
\\
\\
/
3 the discussion the last two days and is aware that this 4
section is weak.
5 CHAIRMAN KRESS:
If you were.:7cerned at all 6
about non-condensibles, would it be in the PRHR?
7 Otherwise, I don't see that it's a problem.
8 ME. LANDRY:
Yes, Dr. Kress.
That was where 9
our main concern was when we raised the question.
Could 10 enough non-condensible gas --
11 CHAIRMAN KRESS:
Get into the --
12 MR. LANDRY:
-- be injected to shut off PRHR 13 n
\\
')
/
14 CHAIRMAN KRESS:
That would be the problem.
15 MR. LANDRY:
-- and prevent the cooldown that 16 the PRHR is there to provide?
And the argument was that 17 non-condensibles don't enter the PRHR until so late in the 18 transient that we're now into the long-term cooling mode 19 and it's no longer important.
20 DR. CATTON:
Is this a conclusion that comes 21 just from looking at the system or actually saying, " Gee, 22 this is when nitrogen will be injected into the system.
l 23 And I already don't care about the PRHR"?
24 MR. LANDRY:
That's part of it, timing on when fs
!s_-)
25 the injection would occur.
And part of it is when they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
h (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 452 l
1 realized from the test program that the gases were l
l 2
entering the system.
l
,s
( t
)
l \\
l
3 DR. CATTON:
I think that's a reasonable l
4 argument, but it should be in writing to be responsive to l
5 that requirement.
And if they do that, then also you can l
l 6
decide whether or not you believe it --
7 MR. LANDRY:
Thank you.
8 DR. CATTON:
-- or we can decide.
9 MR. LANDRY:
Okay.
The third point that I'd l
10 like to bring out is that blowdown hydraulic code must 11 contain Moody critical flow model.
Now, we pointed out in 12 the SDSER that the 0737, II.K.3.30 said that the code must 13 have a Moody critical flow model.
And, indeed, NOTRUMP gy 14 does.
However, NOTRUMP is not applying the Moody critical 15 flow model to the ADS.
16 And we have discussed this back and forth with 17 Westinghouse.
We've discussed it internally quite i
18 extensively.
The position that we are in at this point is 19 that we have the precedent in that the BWRs have an ADS 20 which does not use Moody critical flow and is allowed to i
21 use a realistic flow to calculate the effect of blowdown 22 on the transient.
l 23 DR. CATTON:
Let's talk about that for a 24 minute.
%y recollection of what they did with the BWR is s) 25 they use a model that has been tested against actual plant l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1? RHODE ISLAND AVE., N W.
l (202) 234-4433 WAc.ctlNGTON, D C. 20005-3701 (202) 234-4433 l
453 1
data.
I mean, they take that long SRV line.
It has air s
2 at the beginning.
They treat the heat transfer in it.
t i
\\.
J
'~'
3 It's a rather complicated model.
What Westinghouse has is 4
nowhere near that.
5 So I don't think you can really relate.
The 6
precedence for the BWR I don't believe applies to 7
Westinghouse when you look at the scope of the modeling.
8 And the V and V that has been done by GE for the BWR model 9
is far more extensive.
10 MR. LANDRY:
Ivan, the precedence I would say 11 can be applied because the purpose and the philosophy of 12 the unit, of the hardware is similar or parallel, some 13 term similar to that.
\\/
14 DR. CATTON:
But GE is --
~
15 MR. LANDRY:
Now, the acceptance of it I think 16 is where we're still at in looking at the results of the 17 ADS blowdown and the test program to -- we haven't drawn 18 the conclusion yet.
We haven't stated --
19 DR. CATTON:
I'm trying to help you.
20 MR. LANDRY:
-- categorically it's acceptable.
21 We're looking at the final V and V report to determine 22 that is the justification for use of non-Moody acceptable r
23 for this component?
24 CHAIRMAN KRESS:
What will the criteria be,
,~k
/
(,/
25 that the flow predicted is less than you would get from a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i l
454 1
1 Moody model or --
l l
,e 3 2
MR. LANDRY:
No.
That the calculation as a
- (
'G
\\
3 result of the critical flow model assumed or applied is l
l 4
conservative.
i
\\
4 5
CHAIRMAN KRESS:
Conservative means, then, l
6 less flow then the Moody model or less flow than natural?
7 MR. LANDRY:
Right.
Well, what was discussed 8
yesterday, has it been demonstrated that you get the wc s
9 coolant system response for the model that's assumed for 10 the ADS blowdown.
Would Moody give you a better response 11 for the reactor coolant system?
12 CHAIRMAN KRESS:
Are you going to require 13 them, then, to do two calculations, one with the Moody p.
(V) 14 model and one with the model they had?
15 MR. LANDRY:
They used the Henry-Fauske and 16 HEM models, the submittal.
And we talked yesterday 17 afternoon about doing another calculation with Moody to 18 just --
19 CHAIRMAN KRESS:
Just to compare for a few of 20 the --
21 MR. LANDRY:
To simply den,onstrate that they 22 have the most conservative calculation.
23 DR. CATTON:
But there's another -- can I 24 interrupt just a minute?
,e\\
25 CHAIRMAN KRESS:
Yes.
Go ahead.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
455 1
1 DR. CATTON:
There's another issue here, and I,~,
2 that's this business of compensating error.
I think when
)
J
'~
3 you look at the three paths, the break, the ADS 4, ADS 1 4
through 3, you find a lot of compensating error.
5 The result is that the mass in the system, I 6
mean, it might be a little bit lower, a little bit higher, 7
whatever, but it's really not all that bad because of 8
compensating error.
When you look at the GE system, 9
that's not the case.
They have what I think is a true 10 best estimate prediction of the behavior of their ADS 11 system.
12 MR. LANDRY:
Because they've had a full-size 13 prototype.
(\\
's
/
14 DR. CATTON:
Right.
They even account for 15 extra long lines of SRVs and short ones and bends and all 16 this stuff.
It's built into their modeling.
17 If Westinghouse had that and the comparisons 18 with the experiment showed that the actual flow through 19 them based on the pressure drops or whatever were good, 20 then you could say, "Okay."
21 But that's not the case.
It's compensating 22 error.
They have gross errors, one way or the other, in 23 all of these things.
And they balance out.
And that's 24 compensating error, n(v) 25 And I think there are rules against that.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 l
456 1
Aren't there even some words somewhere about compensating 2
error?
That's a guess.
Anyway.
73 I
3 MR. LANDRY:
I don't remember any words in any j
1 4
of the documentation that discussed compensating error.
5 DR. CATTON:
That's a prol'em because you 6
dor. ' t know the --
7 MR. LANDRY:
We understand your point, Ivan.
8 DR. CATTON:
Okay.
Vou don't know how 9
compensating error scales.
10 MR. LANDRY:
That's right.
11 DR. CATTON:
And that's a problem.
12 MR. LANDRY:
We understand your discussion the 13 last two days, Ivan and Virgil.
l
\\
\\I 14 DR. CATTON:
Okay.
15 MR. LANDRY:
And we're still looking at the 16 final V and V, trying to determine what our conclusions 17 are.
You have to realize that we received the final V and 18 V the same time you did.
19 DR. CATTON:
You worked full-time on it.
20 MR. LANDRY:
We've been working hard trying to 21 get through this thing.
There's a lot of material.
22 DR. CATTON:
I agree.
I agree.
There's a lot j
23 of material.
24 MR. LANDRY:
The integral systems tests.
You
,m
(
)
25 heard today a lot about the integral systems tests that NEAL R. GROSS COURT REPORTERS AND TRANSCHIBERS l
1323 RHODE ISLAND AVE., N W l
(202) 2344 433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
(
457 l
I were being analyzed.
In the SDSER, Westinghouse committed
,-s 2
to reanalyze for both SPES-2 and OSU corresponding tests.
t l
l 3
Two-inch double-ended guillotine DVI, the 4
two-i.nch CMT bala:
line, and the two-inch DVI line were j
5 reanalyzed ror each of the two facilities so that there is 1
1 6
a comparison analysis available.
)
7 And we have to go back through the final V and 8
V now and determine what the results of these analyses are 9
and also to look at the half-inch, the one-inch, and the 10 inadvertent ADS.
We're still going back to look at those 11 calculations to determine how they compare.
12 The comparison on these is important because 13 while the last two days we keep talking about these test (A
k /)
14 programs and how certain functions or certain parameters 15 are calculated and measured are not typical or atypical of 16 AP600, we keep having to also remind ourselves that these 17 facilities were not intended to be simulators of AP600.
18 They were for the purpose of gathering data.
19 Now we have to look at:
How do we calculate 20 those data?
Do we calculate the data correctly?
Can we 21 explain what our differences are?
And can we then 22 extrapolate to the AP600?
23 In that discussion this morning, one of the j
24 points which you brought up was the question of l
,r8
(,)
25 nodalization.
And one of the concerns which the staff NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
458 1
raised with the applicant a number of times has been the fm 2
consistency of nodalization between the different test 3
programs and the AP600.
4 We have reminded Westinghouse several times of 5
the work done for CSAU and the results of the CSAU, which 6
said that you maintained consistent nodalization, which 7
doesn't necessarily mean identical nodalization, as Mike 8
Young, was discussing this morning, but:
Can you justify 9
a different core nodalization?
What's the reason for a 10 different core nodalization in OSU versus SPES-2?
11 We're going through and we're now looking at 12 those nodalization diagrams and the justification for the 13 differences to determine that they have fulfilled what's A
t f
14 called the spirit of CSAU, the intent.
15 CHAIRMAN KRESS:
Is the figure of merit for 16 that going to be the level in the core, minimum level?
i 17 MR. LANDRY:
If the core doesn't uncover --
l 18 CHAIRMAN KRESS:
At all?
19 MR. LANDRY:
So we're --
20 CHAIRMAN KRESS:
I mean, is that --
21 MR. LANDRY:
-- in a Catch-22 there.
The core l
l 22 doesn't uncover.
l 23 CHAIRMAN KRESS:
How close does it come to 24 being uncovered?
Is that --
25 MR. LANDRY:
What is the total system mass, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l l
459 l
l 1
the mass inventory?
Is the mass inventory correct?
Is
,r~S 2
the break flow correct?
Are the timings of events
.Y 3
consistent?
4 DR. SCHROCK:
The way we know that core 5
doesn't uncover is interpretation of experiments, scaled 6
facilities.
So our knowledge that AP600 doesn't uncover 7
depends upon those experiments plus our understanding of 8
the scaling, as opposed to confidence in the computer 9
codes that tell us it's --
10 MR. LANDRY:
That's right, Virgil.
It's a 11 combination.
It's a combination of consistency within the 12 test program, consistency with the code calculation, 13 consistency in understanding the scaling of the p
(
I L
14 facilities.
We're not basing --
15 DR. SCHROCK:
Do you have the same -- if you 16 had to identify source of your confidence in that end 17 result, where would you rank experiment scaling versus 18 code?
19 MR. LANDRY:
Do you want me to speak as an 20 experimentalist or do you want me to speak as a core 21 developer or as a regulator?
I've been all.
22 DR. CATTON:
As a technical person.
23 MR. LANDRY:
Speaking for myself, the basis is l
[
24 all combined.
If I see one that's completely out of line
[Q
^
25 with the others, I'm concerned.
I want to see all l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
i l
460 l
1 combined.
I want to see the test programs are consistent, 2
the scaling is justifiable, and that the code is
(
I 3
consistent in its calculation or I can explain where the 4
inconsistencies are and why going from facility to 5
facility and scaling up.
6 Now, the ideal world would be to have a 7
prototype facility.
8 DR. SCHROCK:
But will you use your own codes 9
to any extent to evaluate this issue --
10 MR. LANDRY:
Yes.
11 DR. SCHROCK:
-- of NOTRUMP validation?
12 MR. LANDRY:
Yes.
Research has been 13 performing a number of assessments, a number of s
_s 14 calculations with NRC-backed codes.
We use that in a 15 confirmatory manner.
We don't want to say audit because 16 we don't try to match numbers.
We use it in a 17 confirmatory manner to support our decisions in a 18 regulatory manner.
19 DR. SCHROCK-Will that information be shared 20 with the ACRS?
21 MR. LANDRY:
Of course.
22 DR. CATTON:
I think we saw a lot of it in Los 23 Angeles, didn't we?
l 24 MR. LANDRY:
Yes, right.
In Los Angeles, you 1
(O) l 25 received a lot of the assessment of RELAP5.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
461 1
DR. SCHROCK:
No.
That's not the intent of my 2
question.
The intent of my question is to examine NOTRUMP 7s t
i n
f 3
using other codes as a yardstick to measure NOTRUMP 4
performance.
5 MR. LANDRY:
If you mean compare number to 6
number, my intent was not to do that but to use our codes 7
as confirming that NOTRUMP is predicting in a reasonable 8
manner.
If our code comes along and says --
9 DR. SCHROCK:
Sort of global bottom-line 10 result that they both say that core doesn't uncover.
11 MR. LANDRY:
If we do a calculation and it 12 says the core completely uncovers and you melt the fuel l
13 and another code says you don't even uncover, obviously k$
14 there's a mismatch.
15 DR. SCHROCK:
I was thinking of a different 16 process in which there may be questions about adequacy of 17 models in the NOTRUMP code.
And can you not form a more 18 intelligent view on that by use of both experimental data 19 available, the test results available, and other code 20 calculations?
21 I mean, the other code calculations give you a l
22 little broader insight inside the models within the code, 23 I think, certainly a different viewpoint of it anyway.
24 MR. LANDRY:
We'll be doing that where we can
/~~N
()
25 with the models which we can because we have to keep in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W (202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234-4433
462 l
1 mind that the NRC codes are not Appendix K models.
We'll i
2 be doing the comparisons where we can do the comparisons,
' iS t
N,)
l 3
where it's reasonable and justifiable to do them.
4 If we're comparing an Appendix K-required 5
correlation with a model that does not use that 6
correlation, it's very difficult to compare and draw 7
reasonable conclusions.
8 DR. CATTON:
But one place you could use it is 9
in questions about predictions of the behavior of the 10 facility, where there -- I don't like to use the word 11
" gross" too often but significant differences.
12 And the argument is that "Well, this is the 13 cause and we can just make a model modification.
And
'wd 14 that's it.
And we'll only do that on AP600."
You 15 certainly could go in with your best estimate two-fluid 16 model or whatever and say, "No, it's not that model.
And j
l 17 it's probably something else.
You've got to fix it."
l 18 Is that the sense you're looking for?
19 DR. SCHROCK:
That's what I was getting at.
20 DR. CATTON:
Yes.
That to me is a good use of I
l 21 the tool you've got.
22 CHAIRMAN KRESS:
Have you not lready done a 23 number of these cases with the RELAP?
24 MR. LANDRY:
Yes, we have.
,a
- i.,)
25 CHAIRMAN KRESS:
So you just have to dig out NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
463 1
the calculations and --
7-2 DR. CATTON:
And look in detail.
I mean, why
(
/
~'
3 this delay in the CMT?
Westinghouse is telling you what 4
they think.
You probably could take a look.
And does 5
RELAP do the same thing?
And if it doesn't, why doesn't 6
it?
And is that the cause?
7 MR. LANDRY:
And that's one of the things that 8
we have plans to do.
To begin with, the gross comparison, 9
"Here's what RELAP says.
Here's what NOTRUMP says,"
10 overall gross picture.
11 Now, are there differercoe.'
Why?
And then 12 start zeroing in on:
What's the CMT doing with the 13 thermal stratification model versus what's it doing
[
)
\\/
14 without?
What's the PRHR heat transfer?
What is it over 15 here?
16 Yes, we do intend to do that.
17 DR. CATTON:
I don't remember the predictions 18 of the integrated ADS 1 through 3 flow being so far off 19 with RELAP.
Well, it might be.
I just don't remember it 20 being that bad.
21 MR. LANDRY:
We haven't sat down and started 22 to do those comparisons yet because we're still trying to 23 wade through tha material we have.
I have done other 24 calculations in the past, not for this specific purpose,
.s q_,)
25 but I've done AP600 RELAP5 cases versus predictions which NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234 4433
464 1
Westinghouse had run using one of their codes and almost 2
overlaid the prediction.
But that wasn't for an Appendix
,f S
'u)
\\
3 K calculation.
4 So that's really not germane as a specific 5
justification for what we were doing, but it's a yes, we 6
have done that in the past.
And yes, we will continue to 7
do so.
8 (Slide) 9 MR. LANDRY:
The status at this point, as I 10 said yesterday morning when we began, the final V and V 11 report is currently under review.
The staff has not drawn 12 final conclusions as to its adequacy at this point.
13 We have said preliminarily that it appears to gg) i t
14 be a very complete document.
There are some gaps, as the 15 Committee has pointed out.
And Westinghouse has heard 16 those complaints and those concerns about where those gaps 17 are.
18 Sometimes we do have those gaps for reasons 19 that I said yesterday.
Some of us, even reviewers, are 20 too close to the work and will have other knowledge and 21 will read the material, not realize it's a gap because we 22 have other knowledge.
And we don't see it as a gap.
23 So the comments of the Committee and the 24 comments of the consultants have been very, very useful j'"g
(,)
25 for us to point out where we may have left something out.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
465 1
I appreciate very much the chance to hear these comments 2
and get this feedback.
It's going to help us quite a bit.
s
/
\\
t 3
The final safety evaluation report is 4
currently slated for the end of August to go to Projects.
5 Based on the discussion in the last two days, we have to 6
also keep in mind that there are a number of things that 7
have been pointed out to us that we're considering.
There 8
are a number of things that we want to go back and look at 9
very carefully.
10 Speaking from the staff side, we received a 11 surprise yesterday in finding out about this lowering the 12 level on the IRWST in the calculation.
I wasn't aware of 13 that.
And now I'm going co go back -- and I've already
(~~\\
u-)
14 talked to the contractor -- go back and we're going to 15 look at the final V and V report with a little more care 16 to find out if there are other surprises for us in there.
17 This wasn't discussed with us.
So I don't care for last 18 minute surprises.
We're going to go back and look 19 carefully now to see that there aren't any others.
20 So that's the status of where we are today.
21 Again, speaking on behalf of the staff, I appreciate the 22 chance to get this feedback from the Subcommittee and from l
23 the consultants.
I think it's been very useful.
We've 24 had some good support from the Subcommittee.
,q
)
25 CHAIRMAN KRESS:
Thank you.
We try our best.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE,, N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
466 1
MR. LANDRY:
Having Dr. Zuber attend some of 2
our meetings, provide us with support, the support, the
,s
)
3 comments from the Subcommittee and the consultants has 1
4 been very, very beneficial.
5 CHAIRMAN KRESS:
At this time, then, I would 6
want to solicit some final quick comments from our two 7
consultants with the admonition that I would like to have j
8 these in writing, but --
9 DR. SCHROCK:
Not only you would like to but 10 you must.
11 DR. BOEliNERT :
We must, 12 CHAIRMAN KRESS:
We must have them in writing.
13 DR. BOEHNERT:
I was going to say we must.
r\\.
t\\~)
14 DR. CATTON:
You have a rather large club.
If 15 we don't send the report, we don't get paid.
It's very 16 effective, very effective.
17 CHAIRMAN KRESS:
You know, the time frame for 18 that is I'd like to have them at least a week before our 19 next full ACRS meeting.
20 DR. CATTON:
Which means mid-August.
21 CHAIRMAN KRESS:
Yes, somewhere in there.
22 DR. CATTON:
That's plenty of time.
23 DR. BOEHNERT:
August 15 maybe?
24 CHAIRMAF KRESS:
Yes.
,.~.
i
)
25 DR. B03HNERT:
Is that possible?
s NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 467 1
DR. CATTON:
Yes.
Well, I'll have it for you 2
sooner.
fs l
l i
\\j 3
CHAIRMAN KRESS:
Okay.
4 DR. BOEHNERT:
Good.
5 CHAIRMAN KRESS:
At this time you're welcome 6
to give any quick reaction comments.
7 DR. SCHROCK:
Before we start that, could I 8
remind Westinghouse that they promised to give me 9
dimensions relating to --
10 DR. BOERNERT.
I think I've got it right here 11 for you.
12 DR. SCHROCK:
You've got it for me?
Okay.
13 Because-I need to have that when I write my report.
t 4
's _
14 DR. BOEHNERT:
Yes.
It's right there while I 15 think about it.
16 DR. SCHROCK:
Thanks.
17 DR. CATTON:
Okay.
Do you want me to just get 18 started?
19 CHAIRMAN KRESS:
Yes.
You start it, Ivan.
20 DR. CATTON:
There are two parts to this 21 process, and we really only heard one of them.
And that 22 deals with the code.
I think the other part should have l
23 been heard first.
And let me try to explain what that's 24 all about.
l x,
)
25 I think Westinghouse first needs to q
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
468 1
demonstrate that the data set -- and this means separate 2
effects and integral testing -- are adequate to accomplish 7x t
\\
/
'~'
3 the V and V task.
And this includes a demonstration that 4
all phenomena highlighted in the PIRT are covered and that 5
the scale is appropriate.
6 This was done by Banerjee with the group from 7
Idaho and presented to us in Los Angeles in April.
8 DR. BOEHNERT:
February.
9 DR. CATTON:
February?
10 DR. BOEHNERT:
February.
11 DR. CATTON:
And also by Wulff at BNL.
I 12 haven't seen their final reports and suspect they're 13 probably still in draft stage, and I would like to see 3(b 14 them.
15 They are two examples of what is needed.
What 16 they were able to do is convince themselves through a 17 combination of scaling and so forth and looking at the 18 data that by appropriate selection, you could develop this 19 complete data set.
And the reason I think it should have 20 been done first is because if it would have been done, 21 then there would have been no questions about the adequacy 22 of the experimental side of what we were seeing.
J 23 We wound up getting tied up sometimes in, 24
" Whoa.
Was this real or wasn't it real?
Is the scaling Ch(,)
25 appropriate?
Is it meaningful to AP600?
I think you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
469 i
l l
1 avoid all of that by doing the other first, and I'm sorry I
2 that Westinghouse didn't see fit to have that meeting g3
\\
i l
3 before we had this one.
It would have gone a long way to
'~
4 shortening up the amount of time we have to spend and 5
reducing the agony.
6 The second part is the code itself.
And what 7
we heard the past two days is only the code.
As it 8
stands, it's unacceptable to me.
There are too many loose 1
9 ends.
Things are not properly explained in the report.
10 Models known to be inadequate are used for expediency.
I 11 think it's a weak effort, and I'm not sure why.
12 If the first part is done adequately, one can 13 argue safety by a combination of calculation and data, as 14 was done by RES for RELAP5.
Now, when RES did this, they 1
15 weren't making an argument for safety so much.
They were 16 arguing for the ade Jacy of RELAPS and trying to show that 17 indeed they had done proper V and V.
18 Nevertheless, the bottom line that came 19 through in that exercise was that AP600 is an extremely 20 robust system.
Basically it's a coffee pot, an old 21 percolator coffee pot.
And as long as you got the level 22 high enough, the core stays covered and it's not a 23 problem.
l 24 But that's not enough for a certification
/'%
i
\\,)
25 effort.
I think that it's incumbent on Westinghouse tc NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
470 l
1 make the same case that Research did for their code.
And 2
this hasn't been done.
( es
\\
i
)
l ' ~'
3 I don't believe that this combination of 4
experiment and the code is really the best approach from a 5
technical viewpoint.
But it's the only way I see you can 6
do it at this point.
7 We I don't believe would have accepted 8
calculations based on RELAPS by itself because there were 9
too many problems and there were too many things that 10 weren't quite modeled well.
But the combination of data 11 that was demonstrated to have been appropriate and the 12 code gave a very convincing argument for the AP600.
13 Tom, I'll write all of this stuff up for you, I~')
5/
14 including some of my concerns about Appendix K.
And I'll 15 get it to you.
16 CHAIRMAN KRESS:
Okay.
I appreciate that, 17 Ivan.
18 Virgil?
19 DR. SCHROCK:
Yes.
I guess I don't have an 20 awful lot to add to what Ivan said.
I see another concern 21 with Ralph's last viewgraph, which indicates safety 22 evaluation report is going to be out the door the end of 23 August.
24 That implies a lot of things coming together (n) 25 here in this next month I think.
And I'm a little unsure e
n/
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
471 1
that that can be done without some compromises that I l
2 don't want to see the staff have to make.
That relates
,g
(
)
~'
3 mainly to the adequacy or the quality level of the 4
validation report as it stands.
5 We discussed a lot of details about things 6
that either we disagree with or things which aren't 7
explained clearly or may not be explained at all in some 8
instances.
For example, I pointed out that there was a 9
statement under the description of the code modeling or j
10 models, section that describes the momentum equation, 11 which indicates that there is going to be an explanation 12 for the deletion of momentum flux terms in Section 1.9.
13 I looked again to make sure that that's what
/
n
\\m /
14 that said.
And indeed that is what that says.
But when I 15 look at Section 1.9, it addresses something else, not that 16 issue.
So if it is someplace else, where is it?
17 And I'd be interested in reading what it was 18 that they originally said that explained how it came to 19 that conclusion.
But I think we've shown now that there 20 are, in fact, some places where that is inadequate.
21 If it really is inadequate, I don't see any 22 opportunity for all of this work to get redone in such a 23 way that the V and V report can be made satisfactory in i
1 24 time to issue a safety evaluation report at the end of l r~N
(,)
25 August.
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON. D C. 20005 3701 (202) 234-4433
l 472 1
So I think there's a problem here which is 2
going to force the staff to make decisions that they ought j
l,-~s t
\\
t.
3 not to be forced to make on the basis of what may still be 4
a somewhat incomplete review of the validation report in 5
order to meet a certification schedule.
6 That may not mean that I would think that the l
7 decision to give certification is wrong because I am 8
convinced that the design is good and that safety of the 9
plant is excellent.
But I do think that we have to have a 10 regulatory process which is done carefully and thoroughly 11 or we're in trouble for the future.
12 So I don't know.
This is not advice I'm 13 giving to the ACRS.
It's just informing you that I have
(~N, Y- /
14 this concern.
I don't see that it's possible to 15 accomplish what Ralph says the staff intends to accomplish 16 and still do an adequate job of reviewing the NOTRUMP code 17 for its adequacy to predict the evaluation model level of 18 performance of AP600.
19 CHAIRMAN KRESS:
Let me ask you a question 20 about that, Virgil.
Let's presume the code NOTRUMP is 21 used part of another code to calculate the flows in the 22 different lines during different times of the various 23 breaks.
24 One could take those flows and make a quick l (/^T l
,)
25 estimate of the magnitude of accelerations and the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
473 1
magnitude. of acceleration terms and compare those with the 2
pressure loss terms and the friction terms.
And one could (p) v 3
do that systematically through various parts and show the 4
relative magnitude.
5 DR. SCHROCK:
Not rerun, yes.
6 CHAIRMAN KRESS:
And then what one would 1
7 probably find out is the place it's really important is 8
during those high flows to blowdown line, where you 9
mentioned.
And one could make an argument there that what 10 this does is the neglect of those terms gives you a higher 11 flow rate out the ADS lines than you would have in 12 reality.
13 You could make such an argument.
And then you
/m I'v')
14 could assess the impact on what the figure of merit is of 15 that being the result of the neglect of those.
Would that 16 be an acceptable approach to you to address this question 17 of the momentum flux terms being neglected?
18 DR. SCHROCK:
I guess I would answer that by 19 saying it would be better than not doing something more.
20 Westinghouse, as I interpreted what Mike said, has a 21 commitment to explain somehow these.
And maybe that's a 1
1 22 route they would take to giving such an explanation.
l 23 But I think if this had been reviewed six 24 months ago, a year ago, and these same discussions had ID
(,/
25 taken place that I would certainly have urged the staff to l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
474 1
require that an improvement in the ADS critica] flow 2
models be made, p) ev 3
I would like to see it get resolved the way 4
you suggested, but I think that would be a better thing to 5
do than to simply listen to an arm-waving argument about 6
it.
That would be some level of quantitative assessment 7
of what's wrong with what they've done.
It's probably a 8
good way to go.
9 DR. CATTON:
I guess, Tom, one of the problems 10 I had in listening to all of this is I really can't pin 11 down where the problem is.
When you focus on the separate 12 effects tests for ADS 1-3, what you find it, well, but the 13 data is not clear that it was properly ranged right.
So
,, \\
/
14 there are pieces missing.
15 Now, what is clear is that under some 16 circumstances, they're off a factor of three on the 17 integrated mass loss.
That's clear.
But what the cause 18 is is not.
19 CHAIRMAN KRESS:
I know.
20 DR. CATTON:
It could be because the separate 21 effects tests are incomplete.
22 CHAIRMAN KRESS:
Those sorts of things have to 23 be explained.
l 24 DR. CATTON:
They may have to do more than r~s k,
25 just talk about them, it seems to me.
There may have to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
f I
475 l
l l
1 be some calculations done, some detailed looks at the --
t l
l 7-2 CHAIRMAN KRESS:
Enough needs to be done to
- ^')
1 t i
3 convince a reasonable technical person that they have the 4
understanding of why they're different, I think.
l 5
DR. CATTON:
Right.
That's right.
l 6
DR. SCHROCK:
Do you think that Westinghouse 7
can resolve what it seems to be committed to resolve --
l 8
CHAIRMAN KRESS:
By August?
9 DR. SCHROCK:
-- and that staff can understand 10 what's in the V and V report and assure themselves that 11 they don't have some other hidden surprises in it --
l l
12 CHAIRMAN KRESS:
I share thic --
i 13 DR. SCHROCK:
-- and get an SER out that you i
(~
1tk 14 can be proud of?
15 CHAIRMAN KRESS:
I share this timing concern, 16 yes.
17 DR. SCHROCK:
Okay.
18 DR. CATTON:
I'm willing to bet it ain't in 19 that report now.
So review of the report and the time 20 spent in review of the report isn't going to help a great 21 deal with some of these questions that we raised.
I tried l
22 to look for these things, and I couldn't find them.
l 23 CHAIRMAN KRESS:
I gather Westinghouse's 24 strategy might be here to demonstrate that they have a way (A) 25 to make this code be conservative from the Appendix K NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 i
476 1
standpoint.
-s 2
DR. CATTON:
That will be --
f 3
CHAIRMAN KRESS:
In principle, that ought to 4
be acceptable.
What do you think it might take to make 5
that demonstration?
6 DR. CATTON:
You sort of get caught.
Because 7
of that meeting in Los Angeles and because of what we 8
know, you're absolutely right.
But now how you do it 9
within this complicated system of knobs and dials I don't 10 know.
So maybe you have to worry a little bit about what 11 they do, which way it's going to go.
And that could well 12 be a very difficult question.
13 That's why I'm -- I mean, I wasn't being smart t
t
's /
14 when I said that it would have been better to have taken a 15 best estimate shot because then I know what to do with 16 uncertainty.
17 CHAIRMAN KRESS:
Following up on that same 18 question again, suppose they decide based on knowledge and 19 intuition things to do to make it conservative, like lower 20 the IRWST level and do things about the pressurizer level 21 and maybe even do some things about the thermal 22 diffusivity problem, numerical diffusivity, and maybe even 23 something with tnc ree.
I don't know.
But suppose they 24 decide on ways to deal with those conservatively.
(q,)
25 And they had some suggestions of things they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
477 1
were going to do.
es 2
DR. CATTON:
Yes.
I I
3 CHAIRMAN KRESS:
Now, you could run the 4
NOTRUMP case as a base case and do each one of those 5
things separately, independently of each other, and see 6
what the effect is and have some figure of merit on what 7
you mean by conservative, which has something to do with 8
tining of opening of the ADS and the system inventory and 9
the levels.
Would that be a strategy that might be 10 acceptable without going through severe code changes and 11 additional tt
.ing and verification?
12 DR. CATTON:
In my personal view, certainly.
13 CHAIRMAN KRESS:
What?
-g]
, )s
\\m 14 DR. CATTON:
Only after I would have on the 15 table this data set that's been demonstrated to be 16 adequate because now --
17 CHAIRMAN KRESE Because these -- okay.
18 DR. CATTON:
See, I would be a little 19 unsettled if you went through all of these arguments and 2u tuning and everything and you're comparing it to something 21 that may not be relevant.
22 CHAIRMAN KRESS:
I have a feeling that the 23 data set that was selected for SPES, GE blowdown tests, 24 and the OSU were selected as being adequate.
()
25 DR. CATTON:
Yes.
But now --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBEPS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
478 1
CHAIRMAN KRESS:
And this was agreed upon to fg 2
some extent by the staff.
(v) 3 DR. CATTON:
That's right.
4 CHAIRMAN KRESS:
You're saying that they need 5
to do more to demonstrate that those selected tests were 6
adequate.
7 DR. CATTON:
Call Banerjee on the phone and 8
ask him what he's going to put in the end of his report 9
because he told us that he would suggest a set.
So did 10 Wolfgang, that they would suggest out of this melange of 11 data what you should select to have an adequate set of 12 data for code V and V.
13 I mean, maybe there's a different one of these p_
l
)
\\'
14 tests that you should have used.
Maybe Westinghouse could 15 just say, " Hey, we've used them all."
In part, that 16 question goes away because I think the two scaling studies 17 demonotrated that that data was adequate.
18 Now, if they're using all of their data, then 19 I guess there's no question, is there?
l 20 CHAIRMAN KRESS:
Anyway, I think we've given l
21 Westinghouse some idea of what our initial reaction is and 22 23 DR. SCHROCK-I'd add one other --
24 CHAIRMAN KRESS:
Okay.
fou have another --
[' 3,)
25 DR. SCHROCK:
-- thing related to my concern NEAL R. GROS.,
COURT REPORTERS AND TRANS,..,lBERS 1323 RHODE ISL/,ND AVE, N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
479 1
about the timing.
And that is the fact that the scaling l
l
<x 2
report will not be reviewed until after this final safety (V
3 evaluation report is written.
If the staff will need l
4 information from that, they will not have it in final 5
form.
i 6
CHAIRMAN KRESS:
I guess at this time I'll ask 7
staff.
If they want to make any additional comments, j
8 they're welcome to.
9 MR. LEVIN:
This is Alan Levin from NRR.
l 10 To address the schedule issue, the August 8 is 11 a target.
It's not etched in stone, with due 12 consideration to many of the comments that have been made 13 while I've been here and I understand were also made while
,_ )
l !
N
14 I wasn't here.
i 15 Whether that date can or can't be met is an 16 open question, but we're not going to shortchange the 17 adequacy of the FSER to meet, to artificially meet, a 18 particular date.
l l
l 19 DR. CATTON:
That's a gcod answer.
l l
20 CHAIRMAN KRESS:
Good answe'.
l 21 At this time I guess I'll ask Westinghouse if i
22 they want to make any additional comments.
23 MR. McINTYRE:
I'd make one comment.
- Ivan, 24 you had said that you would have preferred to have had the IT
(_,/
25 PIRT scaling discussion before this.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234 4433
i 480 1
1 DR. CATTON:
Well, yes because I think it's 2
just an orderly process.
You first make sure that the fy
]
3 data that you're going to use or demonstrate that the data 4
you're going to use was t ate for the task.
That's 5
number one.
6 Once I've done that, now I exercise the code 7
against it.
I can pay attention to the code.
It could be 8
that it's just me, that I'm not going to remember all of 9
these things from today to when I hear about the code, but 10 it seems to me it should be first.
11 MR. McINTYRE:
We thought that last December 12 18th and 19th when we met and went over the PIRT scaling 13 report, that we had arrived at a point that we could go r'N k.)
14 forward and look at the codes.
So we had that meeting on 15 this --
16 DR. CATTON:
I'm not --
17 MR. McINTYRE:
-- on the version of the report 18 that we had at :he time.
And we have known what we did is 19 we have been beavering away, taking the comments that we 20 got at that meeting to revise the report.
I don't think 21 at the time when we talked about scheduling these that we l
22 had planned another meeting on the PIRT scaling report.
4 23 DR. BOEHNERT:
Yes, yes.
No.
That's wrong.
24 MR. McINTYRE:
I know that we have one 25 scheduled now.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
481 l
l 1
DR. BOEHNERT:
Well, it was clear that --
2 DR. CATTON:
Definition of look at the code.
i es s
,\\)
'~
3 MR. McINTYRE:
Okay.
4 DR. CATTON:
When discussion about looking 5
about the code was going on, it wasn't these kinds of i
i l
6 conclusions about the code.
What it was was:
What is l
7 NOTRbMP?
l 8
DR. BOEHNERT:
Right.
l l
l 9
DR. CATTON:
Some of us flat didn't know.
10 DR. BOEHNERT:
Right.
l 11 DR. CATTON:
So the intention was that we i
12 would hear about the inner workings of NOTRUMP, not so 13 much the kind of thing we heard today.
/~'T k -)
14 Now, you jumped the gun in making this a V and s
15 V meeting.
That's the part that's premature, not the l
16 discussion of the code.
Sure, you need to hear about the 17 code.
And you can hear about the code any time, what's 18 the basis of the modeling and these kinds of things, 19 certainly.
20 But you've put on the table the V and V part.
l 21 And it forces you to take this piece of data at face value 22 as being appropriate for what you're doing with it.
And 23 it may not be.
24 MR. McINTYRE:
So you were surprised, then, p
)
25 today that we talked about the V and v stuff?
You were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W, (202) 2344 433 WASHINGTON, D C 20005-3701 (202) 234-4433
482 1
expecting this to be a code meeting?
2 DR. CATTON:
Well, I wasn't surprided today.
fm3 l
I 3
It's when I got all the documentation and I started 4
reading through it and I realized, you know, I really 5
don't have the Westinghouse report that's referenced in 6
that document.
Now, I had it, but it was very old, and I 7
threw it away, 1983 or something.
8 DR. SCHROCK:
Thirteen years ago we reviewed 9
NOTRUMP.
10 DR. CATTON:
Fourteen years ago.
.1 DR. SCHROCK:
Fourteen.
.2 DR. CATTON:
As a matter of fact, NOTRUMP 13 might have even been before that.
So I thought there was O
)
\\~/
14 going to be an updated document or part of the package 15 would include whatever is necessary to do the review, and 16 it wasn't.
17 We got your V and V document.
I don't have 18 the old documentation, at least not that I could find.
19 And I do recollect tossing the earlier one because I 1
l 20 thought:
Why do I have such an old document?
l 21 CHAIRMAN KRESS:
We can probably fix that.
22 DR. CATTON:
You probably could.
23 But that's why it was sort of we jumped right 24 into the V and V without having the discussion of the O(,)
25 code, which is what I thought we had agreed on in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W f202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234 4433
483 1
December.
,q 2
But Tom's free.
See, he didn't agree to t
i 3
anything.
4 CHAIRMAN KRESS:
Well, I didn't have a lot of 5
input into this agenda.
6 Anyway, at this point unless there are further 7
comments --
8 MR. NOVENDSTERN:
Let me say a few other 9
things.
I think we had never intended to update the 10 current version of V and V, the old document.
I think it 11 was an oversight on our part not sending out a copy of 12 that old report at the same time.
13 MR. McINTYRE:
I don't think I would use the g) iC' 14 word " oversight."
We expected that you had it.
15 MR. NOVENDSTERN:
It would have made it -- let 16 me not say " oversight."
It would have made it easier had 17 they had it.
18 DR. CATTON:
This Committee did not review the 19 small-break version of NOTRUMP.
l 20 MR. HUFFMAN:
I believe that's correct.
21 MR. NOVENDSTERN:
That's correct.
22 DR. CATTON:
So the version of the report I 23 had was from a large-break LOCA review of NOTRUMP, which 24 is probably the '70s.
m
/
(,/
25 MR. McINTYRE:
No, no.
NOTRUMP was a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
484 j
1 small-break.
(N 2
DR. CATTON:
What was the large-break model
/
'%,/
3 you had?
4 MR. McINTYRE:
SATAN.
5 DR. CATTON:
SATAN?
Okay.
Then I guess I had 6
a SATAN report that I threw out, too.
7 MR. McINTYRE:
We've had this discussion about 8
not having the NOTRUMP WCAP about three times in this 9
Committee.
10 DR. CATTON:
Yes.
11 MR. McINTYRE:
I'm stunned that you still 12 don't have it.
Can the staff get it?
13 CHAIRMAN KRESS:
Could we?
,.s l
's i
/
V 14 DR. BOEHNERT:
We certainly can.
Can we.
Do 1F you have that still?
I don't have it, but. --
j 16 MR. LANDRY:
I have the only copy in the NRC.
17 MK. McINTYRE:
How many copies do you want?
18 This I can do.
19 CHAIRMAN KRESS:
I don't have a copy of it.
l 20 DR. BOEHNERT:
Why don't you send me ten 21 copies?
22 MR. McINTYRE:
Ten copies.
Sure.
23 CHAIRMAN KRESS:
Ten copies.
24 DR. BOEHNERT:
Why not?
rh
\\
V 25 CHAIRMAN KRESS:
One for each member.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C, 20005-3701 (202) 234-4433
485 1
DR. BOEHNERT:
Yes, whatever.
l 2
MR. YOUNG:
Let me just say one thing.
There l
Q 3
was an attempt in the V and V report, in Section 1, to try 4
to bring forward all the features of NOTRUMP which were 5
important for AP600 so that you did not have to go back to 6
that very old documentation to try to weave together the 7
entire story.
8 That was in summary form.
Perhaps not all the l
9 details were in there that were necessary, but that was i
l 10 the intent of Section 1 of the V and V report.
11 DR. CATTON:
It needs to be a little more 12 substantive.
You went from the sort of full momentum i
l l
13 equation with inertial terms and everything to an equation j
r~%
l t
\\
\\-)
14 without it without even giving a basis.
Just it was gone.
l 15 First I thought it was an error.
16 MR. YOUNG:
We realize we missed the momentum 17 flux story.
18 DR. CATTON:
And, as you read through it, 19 there are a lot of things like that in that V and V 20 report.
There's no root cause kind of evaluation of your 21 results.
You just mention them.
Anyway --
22 MR. NOVENDSTERN:
Let me also thank the 23 Committee.
I think it's clear that you spent some 24 significant time before the meeting reviewing it.
And I L 7-(j 25 think some of the feedback we've gotten to take under NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
486 l
1 advisement is very useful and pertinent.
And we 2
appreciate that and also the time the staff has put into
,cy
'~~
3 this.
4 We've worked with the staff for quite a period 5
of time, two years or three years plus.
And f eedhick 6
we've gotten from the staff in their review, Novak 7
certainly helped.
From what we had originally, here we 8
tie up some loose ends.
And we understand much more I 9
think where they are.
10 As Bob said, you know, perhaps we've all been, 11 NRC and Westinghouse have been, too close to this as we 12 have meshed in this for such a long period of time for 13 things that were obviously to us which maybe did not make
, m 14 it into print.
I'd like to thank again both staff.
l l
15 MEMBER FONTANA:
Are we done?
16 CHAIRMAN KRESS:
We're done unless you want to 17 make some comments.
18 MEMBER FONTANA:
I can't make any better i
I 19 comments than that.
20 CHAIRMAN KRESS:
That's what I thought.
21 MEMBER FONTANA:
Yes.
22 CHAIRMAN KRES::
So I think at this time I 23 will adjourn this meeting.
24 (Whereupon, t.e foregoing matter was concluded O )
25 at 2:40 p.m.)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE, N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
.___.___.,,._.._._..___.___.____...____.__..._m.
_ __. - ~
s O
CERTIFICATE This is to. certify that the attached i
proceedings before the United States Nuclear Regulatory Commission in the matter of:
Name of Proceeding: ACRS SUBCOMMITTEE ON THERMAL
(
HYDRAULIC PHENOMENA (OPEN SESSION)
Docket Number: N/A Place of Proceeding: ROCKVILLE, MARYLAND were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclesr Regulatory Commission taken by me and, thereafter reduced to typewriting by me or under the direction of the court reporting company, end that the transcript is a true and accurate record of th foregoing proceedings.
] }
~
/17/ V '/,
4 1/ s i
"CORBETT'RINER Official Reporter l.
Neal R.
Gross and Co.,
Inc.
I l
I l
l l
i i
s iO i
i NEAL R. GROSS COURTREPORTERS ANDTRANSCRIDERS j
1323 RiiODEISLAND AVENUE,NW (202)234-4433 WAS111NGTON, D C. 20005 (202)234-4433 i
O O
O'&
l i
AP600 NOTRUMP SB LOCA CODE REVIEW j
ACRS/ THERMAL-HYDRAULIC SUBCOMMITTEE L
JULY 29/30,1997 Ralph R. Landry Reactor Systems Branch, DSSA/NRR l
O O
O-t NOTRUMP SB LOCA CODE STAFF REVIEW l
Review Background PIRT i
Component Models j
Phenomenological Models l
Reanalysis of Integral System Tests j
Status - Schedule i
f 2
i I
O O
O-i l
i i
NOTRUMP SB LOCA CODE STAFF REV!EW
Background
Basis for Review: Notrump Code Applicability Document CAD (WCAP-14206) - Inadequate Description of Code Modifications
- PIRT lacked sufficient detail
- Code Hardware Models - cursory discussion
- Code Phenomenological Models - not mentioned Result: Extensive RAls (104)
SDSEP.: New PIRT Commitment of assess phenomenological models Commitment of perform benchmark calculations Commitment to recalculate integral systems tests 3
O O
O-i i
NOTRUMP SB LOCA CODE STAFF REVIEW i
PIRT i
PIRT - Original Unacceptable I
PIRT - As Revised by Westinghouse
- Four intervals: Blowdown, Natural Circulation, ADS Blowdown, IRWST injection Cooling i
- NRC PIRT contained five intervals (W Nat. Circ. = NRC Passive Decay Heat Removal and CMT Drain to ADS Actuation)
{
- All High ranks agree, all Medium ranks in NRC PIRT captured in assessment program Staff Finds PIRT Applicable to AP600 l
t i
4
t i
NOTRUMP SB LOCA CODE STAFF REVIEW Component Models ADS - NOTRUMP tended to overpredict flow rates in OSU and SPES-2 l
- Source: possibly related to characterization of flow regimes in piping
- Final V&V being reviewed for this concem CMT - PIRT ranks flow regime high I
- Concem: lack of thermal stratification
- NOTRUMP predicts proper discharge flow
- Final V&V being reviewed for this concem 1
PRHR/lRWST - Heat transfer correlations modified to include Shah condensation correlation and Lienhard and Dhir correlation -
- Final V&V being reviewed for acceptability of modifications 4
RCP - Assess by benchmark calculatsons
- Final V&V being reviewed for acceptability l
A large number of ADS and CMT component tests were committed to for reanalysis. Results in the Final V&V are under review l
5
i t
STAFF REVIEW l
Phenomenological Models t
Model Modifications Learned in Review of Testing Program Analysis Reports t
18 Models Modified
[
- 3 modifications dropped Assessment of Models Required by Staff
- Resulted in additional model modifications t
i 6
[
t
i NOTRUMP SB LOCA CODE STAFF REVIEW Phenomenological Models SIMARC Drift Flux Methodology Inplicit Gravitational Head i
i Drift Flux Correlations to Apply SIMARC Horizontal Flow Drift Flux Model I
Revised Choking Model Volumetric Flow-Based Momentum Equation EPRI Flooding Drift Flux Model Condensation
}
Cor. tact Coefficients Critical Heat Flux l
Internally Calculated Liquid Reflux Flowlinks Two-Phase Friction Multipher j
Mixture Level Overshoot Henry-Fauske/ HEM Critical Flow l
Fluid Node Stacking Logic l
Bubble Rise Pump Model Transition Boiling Correlation i
i i
h i
o o
o I
i i
STAFF REVIEW 10 CFR 50, Appendix K
. TMI - 2 Lessons Learned Led to NUREG-0611 for Westinghouse PWRs
- WFLASH models considered deficient for SB LOCA
{
NUREG-0737 Further Clarified SB LOCA Requirements
- Section II.K.3.30 required compliance with guidelines in NUREG-0611 l
I II.K.3.30 Requirements j
- Provide calculated validation of core heat transfer and two-phase level
- Validate condensation heat transfer and effects of non-condensible gases j
- Blowdown hydraulic code must contain Moody critical flow model j
l i
l l
I i
[
l 8
1 i
l
O O
O k
i NOTRUMP SB LOCA CODE STAFF REVIEW Phenomenological Models Two-Phase Level Swell
- AP600 SBLOCA not predicted to uncover core
- Two-phase level swell not assessed for low pressure, low flow of AP600
- Tests for assessment l
- G-2 tests: Mid to low pressure, moderate to low power i
i
- GE tests: Mid pressure with/without vessel restriction, break size
- Achilles tests: Integral System Test at low pressure I
9 1
O O
O i
L i
STAFF REVIEW l
Phenomenological Models i
t G-2 Tests: Two powers at each pressure - 780, 400,100, 50, and 15 psia j
GE Tests: Two break sizes at 1,000 psia with/without restrictor plate f
~
j Achilles Tests: Tests at 30 psia and 18 psia
[
i
[
i t
i 10 i
[
L I
O O
O-NOTRUMP SB LOCA CODE STAFF REVIEW Integral System Tests
- Six SPES-2 Tests and Seven OSU Tests to be Reanalyzed
- 2 in. CL breaks for comparison
- DEG DVI line breaks for comparison j
- 2 in. CL/CMT balance line breaks for comparison
- 2 in. DVI line breaks for comparison
- 0.5 in. CL,1 in. CL, and inadvertent ADS-1 i
I 4
k 11 I
i l
l u
1 O
O O-i I
f
?
t i
t i
i NOTRUMP SB LOCA CODE STAFF REVIEW i
f Status Final Validation and Verification Report under review i
Final Safety Evaluation Report end of August to Projects i
I I
f t
I 12 t
t i
i f
-