ML20149E052
| ML20149E052 | |
| Person / Time | |
|---|---|
| Site: | Satsop |
| Issue date: | 01/20/1988 |
| From: | Vissing G Office of Nuclear Reactor Regulation |
| To: | Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8802100196 | |
| Download: ML20149E052 (32) | |
Text
-.
- pa atog'o UNITED STATES
~g NUCLEAR REGULATORY COMMISSION o
- y WASHINGTON, D. C 20555 k***o /
January 20, 1988 Docket No. 50-508 MEMORANDUM FOR:
The record FROM:
Guy S. Vissing, Project Manager Standardization and Non-Power Reactor Project Directorate Division of Reactor Projects III, IV, Y and Special Projects Office of Nuclear Reactor Regulation
SUBJECT:
SUMMARY
OF MEETING OF THE STAFF WITH WASHINGTON PUBLIC POWER SUPPLY SYSTEM CONCERNING THE WNP-3 S0IL-STRUCTURE INTERACTION / DECONVOLUTION ANALYSIS INTRODUCTION A meeting of the staff with representatives of Washington Public Power Supply System (WPPSS or the Supply System) was held at the offices of the NRC in Bethesda, Maryland, on December 15, 1987.
The purpose of the meeting was to resolve some outstanding questions concerning the soil-structure interaction /
deconvolution analysis used in tha WNP-3 seismic evaluation as described in the WNP-3 FSAR. Enclosure 1 provides the list of those who attended the meeting. provides the viewgraphs which were presented during the meeting.
BACKGROUND Brookhaven National Laboratory (BNL) are the technical review contractors on this issue.
BNL provided a report by letter dated October 15, 1987, which indicated that there was not a sufficient basis to conclude the review and prepare a TER. While the applicant has satisfactorily responded to several questions raiseo in a request for additional information of April 1, 1987, there are still some concerns regarding the applicant's deconvolution analysis.
Therefore, BNL proposed a meeting with WPPSS in order to the resolve the concerns raised in the BNL report.
The BNL report was sent to WPPSS on November 10, 1987.
DISCUSSION WPPSS discussed the deconvolution analysis and related issues as indicated in the viewgraphs. A sumary of the main issues discussed at the meeting to resolve the questions concerning the SSI/ Deconvolution Analysis performed by WPPSS for the WNP-3 plant follows:
The applicant's staf presented some data to answer the questions raised by BNL on the applicants response of July 31, 1987, to our request for additional information of April 1, 1987. Specifically the applicant's staff discussed the procedure used to obtain the side wall interaction springs and damping parameters used in the half-space analysis.
The staff indicated that the applicant's methods of determining these impedance parameters were not
!!O Ob O
@0 A
. appropriate. For example, fixing the bottom nodes of the finit: element model for determining the sidewall spring stiffness did not produce the proper stiffness values.
The staff also suggested that assuming the foundat:en to be rigid and obtaining the impedance functions using the procedures described in the currently available literature might produce reasonable results.
Next the applicant's staff tried to show that the response determined by its half-space analyses of Category-I structures, piping, etc. were bounded by the original (finite element) seismic design ro ults.
The staff indicated that the applicant's conclusions appeared to be based on comparison of responses at the ZPA level and that the responses at lower frequences were not included in their presentation.
After a detailed discussion of the various aspects of the applicant's analysis procedures presented, the staff requested the applicant to provide additional information/ justification on the following items:
1, Justify that the ZPA alone represents an acceptable upper bound for the structural parameters of interest for Category I structures.
2.
Justify that the frequency independent impedances are adequate for the frequency ranges of interest at the WNP-3 site for the embedded foundation conditions.
3.
Compare sidewall impedances used in the half-space analysis with those obtained from analytical solutions for these parameters available in the literature.
4.
Provide detailed calculations to show that stresses developed in the steel containment vessel as predicted by the half-space aralysis are within acceptable limits.
The staff informed the applicant that, only after pr)viding satisfactory responses to the above items, a comparison should be made of the spectra from the half-space analysis with the design basis spectra (obtained from finite element analysis using deconvolution).
If the design basis spectra envelope the half-space analysis results, then the method for obtaining the design basis reported in the FSAR will be considered acceptable.
If not, the applicant should consider the specific site effects in determining the ground motion input to be used in the SSI analysis. After that, the staff will perform an audit of the applicant's decenvolution calculations. However, it should be understood that the level of ground motion is subject to modification depending on the results of the Seismology Program review.
In response to the applicant's request concerning the level of detailed analysis that would be expected in justifying the accuracy of its analysis procedure, the staff indicated that the containment analysis should be as complete as possible in order for the staff to resolve this issue.
L
4 In summary, the applicant was asked to review and reevaluate the basic input parameters used in the deconvolution /SSI analysis and then compare the half-space analysis results with the design basis spectra reported in the FSAR.
)
fY=A Guy S. Vissing, Projeg Manager Standardization and Non-Power Reactor Project Directorate Division of Reactor Projects III, IV, Y and Special Projects Office of Nuclear Reactor Regulation
Enclosures:
As stated t
i i
I
ENCLOSURE 1 ATTENDANCE LIST FOR MEETING WITH WPPSS CONCERNING 50Il STRUCTURE INTERACTION /DECONYOLUTION.
ANALYSIS FOR WNP-3 NAME ORGANIZATION Guy S. Vissing NRC/NRR/PDSNP Ramon Pichumani NRC/NRF /?EST/ESGB Carl Custuntino NRC/BNL Doug Coleman Supply System Licensing Dave Bosi Supply System Engineering Mike Hsieh
-EBASCO Civil Robin Wang EBASCO Service Inc.
Kao Ding Chiu EBASCO Civil Choul Ho Yum NRC/NRR/ DEST /ESGB (IAZA)
Goutam Bagchi NRC/NRR/ DEST /ESGB Steve Prussman EBASCO-Licensing A. T. Philippalopoulos BNL/NRC R. Rothman NRC/ESGB,
a
,t P
1 j
WNP-3:
DECONVOLUTION ISSUE 3
0 HISTORY /0VERVIEW D COLEMAN 2
j 0
ADDRESS CONCERN REGARDING DIFFERENCES M HSIEH k
IN SPECTRAL ACCELERATIONS BETWEEN ELASTIC HALF SPACE AND FINITE ELEMENT i
?
ANALYSES.
O ADDRESS CONCERNS REGARDING FOUNDhTION D BOSI j
SPRINGS USED IN THE ELASTIC HALF SPACE ANALYSIS.
)-
O' DISCUSS BASIC PREMISE FOR STRUCTURAL D BOSI
]
AUDIT FINDING #1.
l s
1 b
?.
3 1i i
r.
7 i
?
o 6
i
WNP-3:
DECONVOLUTION ISSUE HISTORY 0
NRC FIRST RAISED THE QUESTION IN REQUEST FOR ADDITIONAL INFORMATION ITEM 220.13 MAY 3, 1983.
O NRC AUDIT FINDING #1 RESULTED FROM THE NRC STRUCTURAL AUDIT CONDUCTED IN NEW YORK SEPTEMBER 1983.
O NUREG-0800, STANDARD REVIEW PLAN, SECTION 3.7.2, REV 1 WAS ISSUED JULY 1981.
0 10CFR50.34(G) REQUIRES THAT "APPLICATIONS FOR LIGHT i
WATER COOLED NUCLEAR POWER PLANT OPERATING LICENSES DOCKETED AFTER MAY 17, 1982 (INCLUDES WNP-3) SHALL INCLUDE AN EVALUATION OF THE FACILITY AGAINST THE j
STANDACD REVI EW PLAN (SRP) IN EFFECT ON MAY 17, 1982 l
OR THE SRP REVISION IN EFFECT SIX MONTHS PRIOR TO THE b5CKET DATE OF THE APPLICATION, WHICHEVER IS LATER".
THEREFORE, WNP-3 MUST PROVIDE AN EVALUATION AGAINST THE SRP IN EFFECT ON-MAY 17, 1982.
O SUPPLY SYSTEM /EBASCO PERFORM AN ELASTIC HALF SPACE ANALYSIS TO DEMONSTRATE THE VALIDITY OF THE ORIGINAL FINITE ELEMENT ANALYSIS (DESIGN BASIS).
THE RESULTS
]
WERE DOCUMENTED IN A SUBMITTAL TO NRC JUNE 27, 1984.
I O
THE RE99LTS WERE DISCUSSED IN A MEETING WITH THE NRC STAFF JULY 10, 1984.
0 THE SUPPLY SYSTEM SUBMITTED FURTHER ANALYSIS ON JULY 14, 1986 PROVIDING FURTHER VALIDATION OF THE ACCEPTABILITY OF THE ELASTIC HALF SPACE MODEL.
i
i l
i i
1
}
O ADDITIONAL NRC QUESTIONS WERE TRANSMITTED APRIL 7,1987,
'i
'I O
SUPPLY SYSTEM PROVIDED RESPONSES TO THE LAST SET OF
)
QUESTIONS ON JULY 31, 1987.
Il 1
9'
'l.
.:j i
a
'}
l 1
1 4
6 1
k 1
i
}
l t
[
)
l l
l l
3A.
THE RESULTS OF THE HALF-SPACE A'NALYS!S CORRELATE WELL WITH THOSE ESTABLISHED FOR THE WNP-3 DESIGN BASES UTILIZING A FINITE ELEMENT APPP0ACH CONSIDEPlflG THAT THE CONTROL MOTION WAS /3DLIED AT TWO DIFFERENT LEVELS; AT THE FOUNDATION LEVEL IN THE HALF-SPACE i
ANALYSIS AND AT THE PLANT GRADE Id THE FINITE ELEMENT ANALYSIS.
THE DIFFERENCES IN RESPONSES BETWEEN THE l
TWO ARE BELIEVED ATTRIBUTABLE TO THE DIFFERENCE IN THE INPUT DEFINIT!0N STATED ABOVE.
l THE PRELIMINARY ENGINEERING EVALUATION OF THE HALF-SPACE RESULTS WAS CONDUCTED IN 1984 FOR A NUMBER OF SAMPLE CRITICAL SAFE SHUTDOWN SYSTEMS WITH EMPHASIS ON THE AREAS OF GPEATER DEVIATION FROM THE DESIGN BASIS RESULTS.
THE RESULTS OF THE EVALUATION ARd l
SUMMARIZED IN ATTACHMENT 1.
i i
l
'l t
y.
2 4
2E. FOR ANY LUMPED MASS-SPRING SYSTEM, THE ASSOCIATED LUMPED DAMPING PARAMETEP.
j WILL INCLUDE THE EFFECTS OF RADIATION (GEOMETRICAL) AND MATERIAL (INTERNAL)
DAMPING.
THE RADIATION DAMPING IN TRANSLATORY MODES (COMPRESSION AND SLIDING) IS ICiOWN TO BE MUCH CREATER THAN THE MATERIAL DAMPING TO AN EXTENT
- j THAT THE LATTER MAY BE DISREGARDED IN THE ANALYSIS.
~i
- i IN THE HALF SPACE ANALYSIS PERFORMED IN RESPONSE TO AUDIT FINDING NO. 1, t
'l THE RADIATION DAMPING ASSOCIATED WITil THE SIDE SPRINGS WAS DISREGARDED 9
9 BECAUSE IT WAS DIFFICULT TO QUANTIFY WITH PROPER JUSTIFICATION.
N USE OF ONLY Tile MATERIAL DAMPING FOR THE SIDE SPRINGS IN THE LUMPED MASS-J SPRING SYSTEM IS CONSERVATIVE SINCE SMALLER DAMPING CENERALLY RESULTS IN HIGHER RESPONSES WITH LITTLE EFFECT ON FREQUENCIES.
g MATERIAL DAMPING OF FOUNDATION MATERIALS DETERMINED FROM TESTS IS GENERALLY y
EXPRESSED IN TERMS OF CORRESPONDING SHEAR STRAIN LEVEL. THE DOMINANT U
STP.UCTURAL RESPONSE IS ASSOCIATED WITl! THE SHEAR WAVE MOTION FOR WilICH THE APPLIED MATERIAL DAMPING IS APPROPRIATE.
1 U)'(
IT HAS BEEN SHOWN THAT RADIATION DAMPING INCREASES WITH INCREASED j
4
[;
EMBEDMENT DEPTH. THE HALF-SPACE ANALYSIS IS CONSERVATIVE IN THIS REGARD
(
SINCE THE HIGHER RADIATION DAMPING EFFECTS WERE NOT CONSIDERED.
v bk
- [ Is *y.f,
- s. / 11 9 !\\ p,ba YQ 1V l}
y d
f 4
t k,
I ll s
ii J
2F. SEVERAL WIDELY ACCEPTED METHODS EXIST FOR COMPLETING SEISMIC DESIGN ANALYSES.
IN BROAD TERMS Ti!ESE INCLUDE IIALF-SPACE AND FINITE ELEMENT METHODOLOGIES.
EXPERIMENTS HAVE SHOWN TilAT THE HALF SPACE MET 110D YIELDS RESULTS JUST AS C00D AS Tile FINITE ELEMENT APPROACH IN CERTAIN CONDITIONS SUC11 AS WHEN THE EMBEDMENT IS LESS THAN llALF OF Tile BUIDLING ALSO KAUSEL ET AL ( } OBSERVED THAT "THE FREQUENCY VARIATION llEIGilT.
OF THESE (HALF SPACE) FUNCTIONS IS NOT VERY DIFFERENT FOR EMBEDDED FOUNDATIONS.
THE MAIN EFFECT OF EMBEDMEN"I SEEMS TO BE AN OVERALL INCREASE IN THE STIFFNESS VALUES OVER THE ENTIRE FREQUENCY RANGE. THIS WOULD ALLOW USE OF Ti!E FINITE ELDIENT TECHNIQUES JUST FOR THE STATIC CASE (TO FIND Tile SIDE SPRINGS)".
FURTHER Y 0 BERDUGO AND M NOVAK( } IMPLIED THAT CONSTANT PARAMETERS (WEIGilTED AVERAGE FREQUENCY INDEPENDENT) YIELDS RESPONSES QUITE CLOSE TO Til0SE FROM THE VARIABLE PARAMETERS (FREQUENT DEPENDENT).
IN OUR LAST RESPONSE, WE STATED THAT OUR HALF SPACE PLUS STATIC SIDE SPRING MODEL GIVES SATISFACTORY RESULTS IN COMPARISON WITH THE FINITE ELEMENT MODEL.
WE BELIEVE THAT FURTHER STUDY IN FREQUENCY DOMAIN IS NOT NECESSARY.
THE SIDE WALLS OF WNP-3 ARE NOT RIGID.
AS E KAUSEL ET AL STATED "THAT Tile ROTATIONAL COMPONENT IS SENSITIVIE TO THE LATERAL SOIL CONDITIONS AND PARTICULARLY TO Tile FLEXIBILITY OF THE LATERAL WALLS.
FOR FLEXIBILE SIDE WALLS THE ACTUAL ROTATION IS SIGNIFICANTLY SMALLER...."
TO DERIVE FREQUENCY SPRINGS FOR FLEXIBLE SIDE WALLS IS VERY CUMBERSOME.
THE EFFORT IS MUCH GREATER TilAN THE ONE-STEP FINITE ELEMENT APPROACH. WIT 11300' WIDE BY 64' DEEP EMBEDDED SIDE WALLS, IT IS VERY HARD TO JUSTIFY THAT THE WALLS CAN BE TREATED AS RIGID.
'2G. SIMILAR TO 2F.
Question No.
220.13 The applicant has allowed a considerable reduction in the esti-(SRP 3.7.1, mated accelerations at the basemat compared to the ' full free-II, and field design response spectrum (over 35% at 10 HZ for horizon-2.7.2, tal SSE).
In the FSAR the applicant is comparing his response 11, 4 FSAR spectrum at the basemat with a 60% design response spectrum.
3.7.1.2)
This is essentially a rock site.
Shear wave velocities of 3000 to 4300 fps were recorded in the supporting medium.
Therefore, it is considered that, regardless of the method used to establish re-sponse spectra, reductions in seismic response at the basemat level would not be expected.
Tnis was specifically confirmed by the applicant in paragraph 2.5.4.7 of the FSAR, the last sentence of which states, "Therefore, tnere will be no amplification in the sandstone, and the caserock acceleration at the site is equal to the acceleration at the plant foundation." The applicant must provide conclusive justification to show that the approach used is technically correct and also will be properly conservative for this rock site.
The seismic analysis methods described in NUREG-0800, Rev.1, Section 3.7.1 and 3.7.2 will be used by the staff as guidance in our evaluation of an acceptable design basis.
Response
On September 26, 1983, members of the NRC Structural Engineering Branch conducted a design audit at the offices of the WNP-3 Archi-tect/ Engineer, Ebasco Services Inc., in New York City.
4 The results of this audit were transmitted to the Supply System
[
via letter from G. W. Knighton to D. W. Mazur, dated November 2, 1983.
The Supply System considers this Question and Audit Finding Number 1,to be essentially the same concern.
Accordingly, we con-sider it appropriate to address the response to this question in l
the context of our response to Structural Audit Finding Number 1.
i I
1 f
Audit Finding #1 In order to proceed with the audit it was decided to discuss with Ebasco the deconvolution analysis for~the plant.
This issue was previously raised in review of the FSAR.
Ebasco has applied the deconvolution methodology to a finite element model of the rock site.
Since the rock is essentially elastic in the range of interest, a reduction in motion at the base mat from that postulated at the surface, would not be expected.
In the analysis, credit was taken for such a reduction and the staff has desagreed since such results would not be expected for a rock site.
Accordingly, the staff will require that the recommendations of NUREG-0800, Section 3.7.2 be followed.
Specifically, either Jn elastic half-space analysis should be performed to ennfirm the finite element analysis, or the site should be considered as a rock site, and a fixed-base analysis should be performed.
In any case the full value of the SSE/0BE should be input at the base mat with-out reduction.
Apolicants Preliminary Response to Findino #1 The above concern was addressed in a letter dated May 3,1983 T. Novak to R. L. Ferguson (Questions 220.13 and 220.16).
The Supply System intends to provide a response to the cited questions by February of 1984.
SGEB Staff Discussion The applicant did not, furnish computations of his deconvolution sensitivity analysis for the staff to examine in the audit (see also audit finding #8).
However, it is considered that the staff direction provided to the applicant during the audit, as described above, represents an acceptable method to verify the adequacy of the structures to withstand seismic (in combination with other) loads.
Any confirmatory analysis by the applicant should include the production of response spectra which can be compared to the applicable response spectra and previous results for the staff evaluation.
3
3 lii 1
s 1
6 Seismic Design for Nuclear Power Plants Robert J. Hansen, editor j
4 j
The htLT. Press Cambridge, Massachusetts, and London, England 4
1 l
l
- It is important to recognize that even with a hard rock site there will be differences in r.otions at tne surface and Figure 28 gives esta frta a 'ite in Japan of an u%erground e
at deptn.
s j
power plant wnere rock extenced to the surface. It was found on a sta-tistical basis tr.st the peak acceleration at ground surf ace was anout
}
twice the paar acceleration at a depth of 300 meters. Thus it is not
.just having a deposit of soil that may cause motions to increase from j
depta up to the surface; tne same trend occurs in rock. #
i L IS9t.M R V Ofe (OM6,RESS t 79 - llO23'1, sant'M. 2A) 4
J f
.i l
I l
l i
r i
)
50 4
e i
4...
s i
4 e
f x
1, o
0 1
=
y e
o N
s d
y u
.h o
m e
I
- j O
l
$ 10 e
~
c e
~
o e
e a
g l
e c
e g
a
=
=w 6-S e
4 a
s I
I f f l E
1 5
10 50 Largest Acceleration at the Base Rock
'5 FIGURE 23 PDK ACCELEMTIONS ON ROCK AND AT 300 METER DEPTH (from Tamuran al. 1959) l i
r e
9 e
9 1
1
_.u
._.. w _ts.. e _... ms -
twaki
!WK G8 earthquakes)ICAL 3 NO.6 e0 20 30 40 30 2
4 6
8 40 2
4 6
8 40 (CLI NO.5 40.61=0 l r
t-20m i
=0.5/40.1-l "a
/ '
MD.4 40.4/40.1-0,
h t -!DM
\\
j l f It i
e NO.3
,3 t-s 30m
'1 an.3ft0.1-I I
(
s e
i 8
NO.2 30.2/40.I' t-200m l
l f i
a
- I i
t NO.e t-330m me.tthD. -
Tomioka NO.6 THK U9 earthquakes)t CAL 1 20 NO.5 40 60 00 ICO 2
4 6
8 10 2
4 6
8
- D (CLI
- 60.6140.I,
^
t-6m 40.5t40.t"
/~,'~~~,
NO.4 Wo.4/40.5-1
['
t-8 00m p
NO.3 NO.5140. l'
[f f
! I t-2s s m a
e a
0 0
0 0
e N0 hg_
f so.tian.i-l i.
",9;b.,,
I
=0.isso.:
Maxirnum Acceleration Magnification Average and Standard Deviation Figure 15 Yertical Distributions of Peak Accelerations at Tomioka and Iwaki Downhole Arrays (after Omote et al.,1984a)
ye1aci(y y,,,nt P. w a y e -- --
per Depth I orn a. Geo-(lon Io91 S*Nav0 Unit yo M t
(m)
(nattet) (9/cml)
No.6 o i
- 3 4
g 5 (GL.)
4 t,y
( 6m)
. i r.~ i _
0,17 i
No.5 1.60 f'"i No.4
(.100 m )
+ ' 00 " ' "n'w 0.10; e
,ag q
<= ~>
% t 88tG
( f t se l 8
tw.
s
- i. i No.3
(.251m) i l
' " ~
, - 3.,.
. ii.
0.5)i t.io i.ro n.,-
i-
' * " ~ '
Vsl Vp V
u,,
(f t-l L.
i 500- s,,.is tr6el 1.20 4.M.
4
- 2.5o i.n a
u s e t~ s sm -
s 1.25 l t,
g,2
(.660s)
+
u--
s t e-100 (ftael awas 2.29
)
r, tw inst.
l.)$j2.60 t
400 -
1.20 I*ee 1.40 %330 2.10
- 4. M t.ee
...~...,,
i 2.30 14.60 2.47 goo.
j Q20 eso No-l e
i si
.+
.<i u 2.a0 s.50 ;
(.950 m )
g,,,,,,
e i.45 1000 cc: 41ter*4 tion of naastw sad ututw Figure 13 Rock Profile and Shear Wave Velocity Profile at Toeloka Downhole Array (after Dinote et al.,1984a) 1
)
)
s 4
+
L g
~
~
s..
Ye1ocit Y weight P. wave pe r Depth Iorwn-Coo-3 'w'ye Unit tion logy volume s
(m) p./gec)
(9N8b f
R$.6 0 ;t i
4 5 6 (CL.)
'i i,.:e
} 0. 9
'T.is';,o 2 es:
l
(.20 m )
+
s No. 's
\\'" **
l?h0 ge.gg i.40 t
- ~
10 -- '
5 5
t e e-a i. =
s t.~
No.4 4 ... H l N *' l l
t.;0 m )
i.a, %
12 r.o. : as i-io0 -
V s '1[V
[tsad.
t is staae l
No.3
(-130m) +,
g 3,
g t.46 t,
s
.l50 -
a l'3.10 s
2.2 - 2.3 y
.g 6ereaise g,,,
3
,g F"d i.52 I
\\
5-l=s44 l
NO.2 s
(-200 m ) + * ", <.,
M.'@...,h.25.[2.65
.i. a.:
u s
i*
( t i ne l.
s t'$3 e
~ '
twt.'
l3.20
,,,,)
( el a-taeneft,33 g,3,oo 8
'l T
250 -
"A g,,,,,,,,,, 3 N,e.~y 1'ts-8.4 t.$0 l3.50 yi q
m, p o,,,,,, q s
s.N 300 -'kA * '* * ' a ** * "
QM L~~1.30 64
]
n==*66t(as) g,g No.1 l'
1.40 !'^
( TIO a ) +
s 2.to tentu
)
.\\
! 8.u - a.s s
M r
1 x
S t,e l te r164 ' El * # 14 A(I tersa.+nd.wd n tene
\\
\\
.y I
i l
.s i
s-i s
t
\\
l s
\\
L a
\\
l s
Figure 14 Rock Profile and Shear Wave Velocity Profile at Iwaki i
isunhole Array (af ter Omote et al. 1984a)
I i
l'
(GAL) Calculated at -21.lm (SilAKE) max. 1.03 1-Ab4 A
i A A1 h$ A $ x x be. A i A-m _.
vpw Ngyy
' y q w y y y
'y y y<
v-'
-1 Calculated at -21.lm y _ (Considering dispersive damping) max. 0.87 it b 4 M k A A A, Auy y,$ i m k
a2 A-m m yy y y y yy vy
, wg y y y y
-l max. 0.90 1
Recorded (-21.lm) bA bA nA
. AA.
v guv r e,.M
.n u,
n_
vy vvy nr v
-1 1
Recorded (-330.8m) max. 0.40 y...',"yf'y Q,-(V ) y ( A '^ / y j'y^,^yM,y'
>^
v.
y y,
-1 Input motion I
I l
l l
I 20 21 22 23 24 25 TIME (SEC)
MAX. ACCELERATION AXPLITUDE (cat.)
0.0 0.5 1.0 1.5 0
\\
.i 1
l f.
100-T 1
l g
200 -
....?:
.) --~. Calculated (SHAKE) 300 -
Calculated (Considering dispersive dampisig) l Figure 17 Comparison of Recorded and Calculated Motions (MS Components). Iwaki Downhole Array (September 13. 1981 Earthquake) (0 mote et al., 1984b)
1 E
l
?
3 5
'l
(
N o. 6=
i1W)[E GL-O N o. 7
-20m-N o. 5
- 7 0 m-N o. 4 j
- 13 0 m-ON o. 3 i
5-2 0 0 m-ONo.2 C
l 4
}
-330m-O N o.1 i
AWL A e d.* er e f'R t tt Rs.,o-4 L f.s r
. o. t l
e 5-t 7 (CL 4.8ml/ml(Cl*W mi a 4 (&L 4.6ml/ml(Cl*WE) 5 (SL-Mal /mlist awn)
)
4 4 (6L-Mal /eal(El.Wm) j
,y 3 (Malidal/eal(Gl*Wm) 9 an 2 (lletWal/El(LL*W E )
1 E3-i
~~~~ /,
i
!j' N p ~ ~~
\\
f ".
- bWAA f
{ ~ ' y n I '_ /
~ ~ '.
=
l N'
%,,/ _ '.~%:V O
0.02 0.05
- 0. 1
- 0. 2
- 0. 5 1
2 5
10 Pers.4(sec) i A v e em.3. it4 4 i.
.f' Spm 4 r.t. A c ce le cat ie n
"[ w ki b.wnh.in A c c. y ( O
,4 g/.t.,, :v 9 1
)
w I
CONCLUSION WNP-3 DESIGN BASIS METHODOLOGY FOR ESTABLISilING Tile FACILITY RESPONSE SPECTRA (I.E. FINITE ELEMENT ANALYSIS) MEETS Tile REQUIREMENTS OF GENERAL DESIGN CRITERION 2.
Tile WNP-3 DESIGN ANALYSIS PROCEDURES PROVIDES AN ACCEPTABLE BASIS FOR THE SEISMIC DESIGN.
AN ELASTIC HALF SPACE ANALYSIS ~ WAS PERF01U4ED WilICH INPUT Tile FULL VALUE OF THE SSE AT Tile BASEMAT.
Tile RESULTS OF Ti!E ELASTIC HALF SPACE ANALYSIS CORRELATE VERY WELL WITH THF. BUILDING RESPONSE SPECTRA GENERATED BY Ti!E DESIGN BASIS FINITE ELEMENT METHOD.
TilERE ARE NO GROSS DISPARATIES BETWEEN THE TUO SETS OF RESULTS.
BOTl! TECTINIQUES INCORPORATE SIGNIFICANT CONSERVATISMS IN MODEL INPUTS AND EXECUTION TECl!NIQUES.
COMPOUNDING THESE SEISMIC ANALYSIS CONSERVATISMS ARE SUBSYSTEM ANALYSES WilICH CONFORM TO RESTRICTIVE DESIGN CODES AND STANDARDS.
Ti!E FINITE ELEMENT METi!0DOLOGY NECESSITATED Tile USE OF DECONVOLUTION ANALYSIS.
DECONVOLUTION ANALYSIS IS AN ACCEPTABLE TECilNIQUE EVEN FOR ROCK SITES SUCH AS WNP-3.
IN-SITU SEISMIC DATA DEMONSTRATES ACTUAL ATTENUATION IN SEISMIC RESPONSE WITH DEPT 11 CONSISTENT WITH THAT PREDICTED BY THE SHAKE MODEL RESULTS.
IT WOULD BE INAPPROPRIATE TO APPLY THE PRESENTLY DEFINED SURFACE FREE FIELD L
SAFE SHUTDOWN EARTHQUAKE AT Tile FOUNDATION (BASEMAT) WITil0UT REDUCTION.
i 1
i 1
.; i e
d 1
REFERENCES 1:
r
- 1) T MASA0, M !!IRASAWA, S YAMANMOTO AND Y KOORI, "EARTHQUAKE RESPONSE OF NUCLEAR REACTOR BUILDING DEEPELY EMBEDDED IN SOIL", 4Til SMIRT,- 1977.
e j
2)
E 1(AUSEL, R V WHITMAN, J P MORRAY AND F ELSABEE, "THE SPRING METHOD FOR EMBEDDED FOUNDATIONS", NUCLEAR ENGINEEERING AND DESIGN 48 (1978).
i I
3.
Y 0 BEREDUGO AND M NOVAK, "COUPLED HORIZONTAL'AND ROCKING VIBRATION j
OF EMBEDDED FOOTINGS", CANADIAN GEOTECHNICAL JOURNAL 9,1972.
4.
NUREG/CR-3805, "ENGINEERING CHARACTERIZATION OF GROUND MOTION'l, VOLS 3 AND 4.
I 5.
C Y CHANG ETAL,. "USE OF OBSERVATIONAL DATA IN EVALUATING THEORETICAL
{
?
MODELS OF GROUND RESPONSE" EPRI WORKSHOP ON EARTHQUAKE GROUND MOTIONS l
IN THE EASTERN U.S., MARCH 31 To APRIL 2, 1987.
n 6.
R J HANSEN, "SEISMIC DESIGN FOR NUCLEAR POWER PLANTS" MIT PRESS CAMBRIDGE, MASS., LIBRARY OF CONGRESS NO. 79-110237.
1 i
i l
i 1
s
?
'i l
1 i
I i
(
)
[
l r
I 9
h F
)
i ATTACHMENT 1
)
e i
1' I
l l
i l
'i I
.._.._._...___.m__...mm.
.m_.__...
..__..__._.m.__
BASIS FOR. ENGINEERING EVALUATIONS OF ELASTIC HALF SPACE RESULTS ASSESS AREAS OF MAXIMUM DEVIATION FROM DESIGN BASIS SPECTRAL RESULTS SAMPLE CRITICAL SAFE SHUTDOWN SYSTEMS MAJOR STRUCTURES PIPING AND SUPPORTS ACTIVE VALVES ROTATING EQUIPMENT HVAC ELECTRICAL POWER SUPPLY INSTRUMENTATION AND RACKS ASSESS IMPACT OF ABSOLUTE MAXIMUM SPECTRAL ACCELERATIONS l
4 i
o.
CATEGORY I ' STRUCTURES O
BY INSPECTION MAXIMUM ACCELERATION LOADS ON MAJOR CONCRETE STRUCTURES ARC BASICALLY BOUNDED BY THE ORIGINAL DESIGN BASIS 0
INCREASED MAXIMUM ACCELERATION OF THE STEEL CONTAINMENT VESSEL IS ACCEPTABLE BASED ON REVIEW OF THE CHICAGO BRIDGE &
IRON CO. STRESS REPORT (IE, DESIGN BASIS STRESSES ARE WELL BELOW ASME ALLOWABLES) 4 e
2
)
-......u.--
u...
..n..
PIPING EVALUATIONS 0
CONTAlunENT SPRAY HEADER O
AUXILIARY FEED WATER SYSTEM 0
Hisu PRESSURE SAFETY INJECTION SYSTEM RESULTS:
O ALL PIPING STRESSES REMAIN WITHIN ASME CODE i
ALLOWABLES 0
SUPPORT LOAD ALLOWABLES NOT EXCEEDED i
1 l
EQUIPMENT EVALUATIONS 0
PRESSURIZER RELIEF VALVES 0
HVAC EoulPMENT/AC ANo CU UNITS 0
HPSI Pune ANo NToR RESULTS:
IN ALL
- CASES, VENDOR INTERFACE REQUIREMENTS WERE ME! (6-LOADING, FREQUENCY, N0ZZLE LOADS) i i
j j
1 l
i
{s ie y
y v
s g
ie ia ]
g a
3 i
^
WNP-3 REACTOR AUXILIARY BUILDING SSE EAST WEST 3 PERCENT DAMPING j
L N0DE 25 EL 416.50 CONTROL ROOM AC UNIT 2A r
TEST RESPONSE SPECTRA (TRS) i 6*0
~
~
- TEST N0.11 F-B/V g\\
~ TEST N0. 20 S-S/V TEST NO. 28 S-S/V N
l l
g i
g s
p g 4.0 i
5
)
y u
i y
i 1
}
l I
2.0
-~
FINITE ELEENT ff" _
EI).STIC HALF-SPACE
=
4
~
FLOOR RESPONSE SPECTRA
' * * ' ~
~I
- ~ ~ - - - '
I ~ '- '~~'~ ' E -
-I-O0
- - - ~
l 0.5 1.0 2.0 5.0 10.0 20.0 50.0 FRE0bENCY (CPS) 2 l
x mamz,.s a-,,e
-mza am.a
w s-----
-sa,.,2
-m s
=
-m
*-w w---
h a
4 D.
$ 3'N i
l
$mvg a
, a bh N >~
E$W cm r
)
'l FN y
i t:
s g
x, e
5 i
32 m.$
j 8
- 5 e
t.h Ahh 5
h i
e--
q_,
g
$$k k
_kh W
ers ass E * :*.i mum
_gi g
W'W W Il i
g e
=-
w 9 5m sE
~
sn I
A l
l l
l a
l A
a A
A e
A 9
9 9
a e
n a
(9) WH1YW373DY
i ELECTRICAL /I & C 0
ALL CLASS IE 480V POWER CENTERS 0
INSTRUMENT RACKS AND PANEL MOUNTED HARDWARE 1
e TEST RESPONSE SPECTRA COMPLETELY ENVELOPES RESutTS NEW SPECTRA RACKS ARE RIGIDLY DESIGNED
+
e
,,__,_gy,,A,,
-A-A o-->
-a s
~-a-A 6
A
-~a=--A---
e 9
,g 1
y g
g 1
f L
$ ll5 dl
- &, c l
h_ -
I
'Y
- R
/
i t
es=
mw 1
go A
i
\\
d
/
C'
?
)
S i
/,
A tw g
h Be".,m k'
IE as I
M i
(
s i
'h liisii I) g
! Es
- E $5 g W-[
b(
4
.N 3
14 N.
T 5M 8 i
=
g-R I<
- llEe 3N 9
- kgg a
.=
1 1
A 1
1 1
1 9
e e
O O
O b
5 0
G) 14011VITGC%
i
~..
i 3
,.y-----
WNP-3 REACTOR AUXILLARY BUILDING SSE VERTICAL 4 PERCENT DAMPING 6.0 N0DE 26 EL 389.00 9
480V POWER CENTERS Q
5.0 TEST RESPONSE SPECTRA (TRS)
T TEST NO. 9 F-B/V x
94.0 TEST NO. 20 S-S/V 8
i l
3 3.0 i_
O u
~
t FINITE ELEMENT
~
ELASTIC HALF-SPACE f
2.0 FLOOR RESPONSE SPECTRA j
~
u 1.0 j
Z t
L
,o2 i
I-i
.- I
,1
___ i 0.5 1.0 2.0
- 5.0 10.0 20.0 50.0 4
FREQUENCY (CPS)
e _
January 20, 1988 i
Y.
In summary, the applicant was asked to review and reevaluate the basic input parameters used in the deconvolution /SSI analysis and then compare the half-space analysis results with the design basis spectra reported in the FSAR.
original signed by Guy S. Vissing, Project Manager Standardization and Non-Power Reactor Project Directorete Division of Reactor Pr0jects III, IV, Y and Special Projects Office of Nuclear Reactor Regulation
Enclosures:
As stated DISTRIBUTION:
"'153EEEE7113?',s,_
t
~>yc. w +.
PDSNP Reading EHylton LRubenstein GVissing OGC-White Flint EJorden i
JPartlow NRC Participants l
ACRS (10)
TKing E
o PDSNP' /
P05 GVig8 LR stein Ol/br/B8 01/4b/88 i
i f
t h
t
\\' f i
I i
l t
, _ _. - -.