ML20128H324

Forwards Responses to Open Issues Documented in Sections 2.1 & 2.3 of Encl 2 to NRC 921113 Summary of 931012-15 Meeting, to Support Accelerated Advanced BWR Review Schedule

ML20128H324
Person / Time
Site:	05200001
Issue date:	02/09/1993
From:	Fox J GENERAL ELECTRIC CO.
To:	Poslusny C Office of Nuclear Reactor Regulation
References
NUDOCS 9302170007
Download: ML20128H324 (125)

Text

':i % V

• 2 :

a p[

- -- y

.x

).

> ~

GE Nuclear Energy l

~

Ker; ocam arm ns cmv re:we. wtae u w:s

^

}{

February 9,1993 Docket No. STN 52-001 L

l Chet Posiusny, Senier Project Manager -

Standardization Project Directorate Associate Directorate for Advanced Reactors L

and License Renewal Office of the Nuclear Reactor Regulation W

Subject:

Submittal Supporting Accelerated AHWR Review Schedule Dear Chett Enclosed are GE's responses to the open issues documented in Sections 2.1 and 23 of

-1

- Enclosure 2 to NRC's letter " MEETING

SUMMARY

OF OCTOBER 12-15,1992," by C, Poslusny, dated November 13,1992. The following two Sechtel reports are included for responses to two of the issues: (1) A Study of Effects of Out Of Phase seismic Shear Forces on the Design of Upper Structures of Reactor Building, and (2) A Study of ABWR Soil Spring Stiffness.

Please provide copies of this transmittJ to Tom Cheng and Gautam Bagchi.

Sincerely,

+

h U

':! fick Fox

' Advanced Reactor Programs 1

9

.. ce;. Gary Ehlert (GE)w/o Enclosure Norman Fletcher (DOE) w/o Enclosure Ting Yu Lo (LLNL) w/ Enclosure s

y

/y2 p.

L JL

1. 'g &

3 p'sont7ooot,3oso, anocx oseooooi-

,/r he w

=

a w.

2.1-Seismic Re analysis of Category I Structures (1)

GE should verify th~at the rock and hard rock site condition (shear w, e velocity.

equal to 20000 ftisec.) is the bounding s,ite condition.

Response

A fixed base time history analysis of the reactor building was performed using the computer program STARDYNE. To simulate the upper bound soil case (RIU) as in SASSI model, the reactor building nodes correspondir.g to perimeter walls below the grade level werc fixed in the STARDYNE analysis. The results of -

analysis are compared with SASSI R1U (rigid soil, uncracked condition)in figures 1 through 3 at key locations in the reactor building in X, Y, and Z-directions.- As shown, these results are in good agreement. Minor discrepancies in the results are due to different methods of analysis and handling of material damping in SASSI-and STARDYNE methods These results confirm that the soil case with Vs=20,000 fusec. is the bounding case.

I J'

ACCELERATION RESPONSE SPECTRA 02 i

... i i SASSI(R10X)

ADWR REACTOR BLDG.

- - STARDYNE NODE 31 X RPV/MS N0ZZLE 22 DAMPING cn 6.

en tn z

O s

F-4 s

trw

__s 4-w U

U 4

f i

'.)

s g

(r F-.

g k Aj W-g 2.

v

' ' ' '~'

O, 2

10-8 10e 108 10 FREQUENCY-CPS FIGURE la

a ACCELERATION RESPONSE SPECTRA 8.

,,iiiia i

e i

i i i i i i

i i

i i e i i SASSI(RIUY)

ABWR REACTOR BLDG, STARDYNE NODE 33 Y RPV/MS NDZZLE 22 DAMPING.

cn s

6.

a m

z

.. o F

/

f.

E

[~

.m J

4-w u

u.

?

J<

E

. F 2

- o k A, i

n.

2.

'y y

m x.....

1 2

.10-1 10e.

10 -

10 L FREQUENCY-CPS FIGURE lb

• p.,

v Mr q:

N ACCELERAT10N RESPONSE SPECIRA l

4.

i i

i i i i i:

i i

a i i i

i i

i ri e i i ABWR REACTOR BLDG.

SASSl(R102)

PJUDE 33 2 STARDYNE RPV/MS N0ZZLE 22 DAMPING cn 3.

l to tn Z

\\

O s

t-<

m tu I

J p-tau U<

J m

l-U ta U-I' fl tn i

.._,,m 1

... m 2

o.

8 10 10-1 10 10 8

FREQUEtJCY-CPS FIGURE Ic f

4 v

r ACCELERATION RESPONSE SPECTRA a.

i i

i i i....

. i i i.

i i. i i. > >

. SASSI(R1DX)

ABWR REACTOR BLDG

- - STARDYNE.

NODE BS X RCCV TDP 22 DAMPING i

cn i

6.

re nn z-o H

- t--

'tr w

.i w'

A-

~

N U

U<

.Et 2

F-u w-j

[L 2.

en h

1 e

O.

10-1

~100 2

102

10 '

[

-FREQUENCY-CPS.

FIGURE 2a:

1 3

gs,

. < - +..

ACCELERATION RESPONSE SPECTRA a.

> >..i SASSI(R10Y).

ABWR REACTOR BLOG.

- - STARDYNE, NODE 89 Y RECV TOP 22 DAMPir4G

- f 6.

- to tn z

O..

s F-4

-w J

4.

w.

U U<

J 4

J, K

y F- --'

uw

- n. -

2, en

' 10-3 '

10e

' ' '. '1 0 '

O.

2 10 FREQUENCY-CPS FIGURE 2b' Y

L

-i ACCELERATION RESPONSE SPECTRA 0.

i i

i i i i ir i

i i

i i i i r i

i > > i i SASSItR102)

ADNR REACTOR BLOG.

STAROYNE t4DDE 09 Z RCCV TOP 22 DAMPIf1G cn i

G.

sn US Z

o s

t-<

tr u.s 4.

u u

,]

)<

tr

(~.

F-U

(.,'

11J

[L 2.

/

i_L.Li _.

i i

e i i 2_1_1_

i i

a s e a i n

i i - a g,

0 8

2 10-3 10 10 10 FREQUENCY-CPS FIGURE 2c

t ACCELERATION RESPONSE SPECTRA 12 i

i i i ii i

i i

.iii i

i i

i i...

t SASSIIR10X)

-4

-- STARDYNE ABWR REACTOR BLOG.

NODE 95 X i

R/D TOP 22 DAMPING 10.

(

c.

I so tn z

B-U s

}--

m 10 Y

tu' 6-Uu<

d

-I i.!.

4 g

4-

)_

it :

u I :

to n.

tn 1

g, 10'3

'10e 102 102 i

FREQUENCY-CPS FIGURE 3a

ACCELERATION RESPONSE SPECTRA 12.

r i

i i i i ir i

i i

i iiir i

i i

ii i i r" SASSl(R10Y)

I AOLJR RE ACTOR BLDG.

STARDYNE

}

NUDE S5 Y R/D TOP 22 DAMPit4G 10.

~ ~

cn I

ro tn z

6.

o s-e H<m tu J

6.

g o

U<

J

• b m

4-gw' ru to -

Q.

tn 2'

0.

~

1 10 s

10 10-8 le FREQUENCY-CPS FIGURE 3b L

1 l

v q

8.

ACCELERATION RESPONSE SPECTRA SASSliR102)

-- - STARDYNE ABWR REACTOR BLDG.

NODE 95 Z R/B TOP 22 DAMPING I

<n s

6 '.

.1 e

u)

Z O

s F

4 m

W J

4.

-t g.

'O

[

U e

N.

.3 J

'I E

V U

n_ i w

p

. w 7

F I

.)

O '.

' '. '1 0 -

10~ 8 -

.10e i

8 2

.10 FREQUENCY-CPS

. FIGURE 3c:

, _ _ +

4 x-.u 4.,4..>

..~4.4 e

g.-

.Lp-.+..+.;

.A e

4 n.+=4

..:9_..w_..

4w.e a

1, s.+<

' (2)~

For demonstrating the adequacy ~ of the dynamic.model, GE needs to clarify and provide the final designs regarding the stn.ctural connections between the sub-compartment _ walls and the RCCV ~shell.

-Response:

The reactor building walls are designed with a gap betiveen the RCCV wall and -

the partition walls of the reactor building. This gap is 'to be filled with a rubber like :

~

material to provide a smoke, fire, and flood barrier for divisional separation. This will be similar to the Mark 11 containment designs.

b h

)

Y t a

.c h

k 4

9 14my e

a (3)

GE should consider the effect of the basemat flexibility in:

(a)

- Stmeture-to-structure interaction analysis (b)-

S ASSI model for individual building analysis.

(c)

Connectivity of the model sticks to the basemat

Response

In order to evaluate the efTect of basemat flexibility, soft soil UBID150 was used to re-analyze both the reactor building and the control building using concrete properties of the basemai. The soft soil profile was selected since the effect off basemat flexibdity is expected to be relatively less significant for the stifTer profilesL In the SSI models, the stick models were connected to the basemat at their respective footprints. The analysis was performed in the X-and Z-directions using the 3D models of each building.

The results of the flexible basemat case (UB1DlF) are compared with the results of the rigid basemat case (UB m" a tigures I and 2 for the horizontal X-and vertical Z shakings for the reawi building As shown in these results, the basemat flexibility etTect generally reduces the horizont'ai responses and increases the vertical responses. The differences, however, are not significant except at node 95 -

in the X-and Z-directions where the difference is relatively larger (see figure Ic and 2c) However, when these responses are compared with the upper bound case results (RlU), as shown in figures 3a and 3b, the flexible basemat. case results are well within the upper bound case results, For the upper baund case, the flexib.c basemat case is essentially the same as the rigid basemat case due to high stifTness of the supponing soil medium.

Since the enveloping results adequately cover the variation due to basemat flexibility effect, no modification of the reactor building results using rigid basemat -

assumptions is considered warranted:

The results for the control building are shown in figures 4 and 5 in the horizontal b

X-and vertical Z-directions, respectively. As shown in these restilts, the effect of basemat flexibility on these responses is triviale l-f l-l'

'l II I

ACCELERATION RESPONSE SPEClkA 4.

a i i ir i

i i

a 7m r i

a i

i i rw ABWR REACTOR BLOG.

UO101X(RIGID DASEMAT)

NODE 33 X UB101FX(FLEX BASEMAT)

RPV/MS tJ0ZZLE 22 DAMPltJG cn s

3 n!

tn Z.

O s

F-<

{

T l

tij La uo 4

_.s

{.

m

\\

d N

\\

\\.

F-g G

'1.

~. -

v j

g i

i i

e e ii i

i i

i e i ir i

i i i i

2_L, 2

10 10-3 10e ici FREQUEtJCY-CPS FIGURE la

AE E E L E R A T 10t1 RESPONSE SPEEIRA 4*

i

~T i

i -i i iT i

i i

i e i F r-~

i 7

i e

i i e i UB101X(RIG 1D BASEMAT)

ABL4R RE AC TOR BLOG.

UO101rX(FLEX BASEMAT)

NODE 89 X RCCV TOP 22 DAMPING ca i

3 to LD Z

O e-e F<

m W

~

J p-w

.3 U

J<

~

x F

U w

s.

)'

tn a

f

'g

,,, m 2

o.

10-2 10e 10 10 2

FREQUENCY-CPS FIGURE lb

ACCELERATION RESPONSE SPECTRA s.

. i. > i.

i i i i. i,

UB101X(RIGID BASEMAT)

ABWR REACTOR BLDG.

- UB101FX(FLEX BASEMAT) f40DE 95 x R/8 TOP 22 DAMPI t4G 4.

... cn i

fa m

Zo tr t--

3.

.:l.

4 oc

. LU

'J w

.f:

u o

f g

.2.

J

.4 i

cr.

.. t-u t-W i

1 a..

h7 /

.i\\/..Ak7

- f,.

\\e;..

.0?

102 10 10e 101 FREQUENCY-CPS:

+

'. L

. FIGURE Ic' t.

.[

t y

y,

+i.s.+.

+,,

-.%g

-, - e 3

y

-u.

w

.,y.#

rp

4 ev' ACCELERATION RESPONSE SPECTRA 2,0 t i

i. i i i t

i i

i iiiia i

e i

i i i i i 1

' t191012(RIGID.8ASEMA T)

-UB101FZ(FLEX BASEMAT)

ABWR REACTOR BLOG.

tJDDE 33 2 RPV/MS 74DZZLE 2r DAMPIf4G

=

A i

1.5 m-y m

'g z

i o

s-

- F--<

tr w

d 1.0 u.

U E

J<

Ii E

t-Y^

- u w.

i (L

'5 I

i s

.0 10-1'

' 10e-8 10 -

102 -

FREQUENCY. CPS

~I

~ FIGURE 2a.

Ji 1-e ACCELERAT10N. RESPONSE. SPECTRA 3.

i 1

e i.

3.

i ii-1 i

i i i i ii i

e i

i i.. i i I

UB1012(RIGID BASEMAT)'

ABWR - RE ACT')R BLOG.

08101FZ(FLEX BASEMAT) t400E 83 Z

- +

RCCV TOP i

22'OAMPlf1G 1

cn

,g.

' f5 ul

z 2i

' O

.+-<

.p 4

..ew

.u -

to.

.u-u 1

e

.s

\\

3 i..

A m

i-V p..

u y

+

LU i

~,

g.

tn

.1 3.

s O'.

2 10'3 10e.

103 10 '

FREQUENCY-CPS FIGURE 2b-7 p'

4

--7 r

y-j qg

'" +

.g T

S ye

<a w

.w W,___

8

- z=.

nsa1 a

--4

m--WJ "

?---=-

-A S

i_

,a 70 i

i 1'

a i

e G

D LB R

i O

T G

CZ N A

I ESPP RSOM TA RE D WDB A

BO/2 R

ANR2 T

8 C

E y

0' P

1

.~

S i E i S e

~. '

N S

O i

.P P i C

e S

E Y

R i C

c N

2 i

E N

U E

R Q

O i i

E G

v U

I j

e R

I T

F F

AR E~

L F'

E

~i e

0 l

i 1

C 4

A

) )

i TT AA MM e

EE i SS i

AA BB

+

i DX IE i

GL IF i R(

( 2 i

2F 11 00

./e i1 1 BB UU w

_ 0 3

2

. g

, 1 e

n nn zomH<tWaWuU4 r

A(tpUWnn EI

- r i

t t

i;

ACCELERAT10N RESPONSE SPECTRA 10.

i i i e ir i

> > i..

R1HX ADHR RE ACTDR BLDG.

118101XtRIGID BASEMAT)

NDDE SS X'

- - 001DIFx(FLEX BASEt %I)

R/8 IDP 22 DAMPitiG-6, cn Ia en l

Z 1

o s

F 6.

m tu hJ tJ U<

l^

A J

i l

ta 5_ t 2.

3 I.

(/< p..Ag p '

~

o i_

y N _ _____ _____

0, 8

2 10'1 10e 10 10 FREQUEtJCY-CPS FIGURE 3a

{

'-t t

. -.a. -.

m ACCELERATION RESPONSE SPECTRA

~6.

.. i i >

R102 ADWR REACTOR 8tFG,

- UB1012(HIGID BASEMAT)

NODE S5 Z R/B TOP

- - UBIDIFZ(FLEX BASEMAT) 22 DAMPING as i

10 ut 2

A.

O N

1-tr tu J

tou O<

J<

l

?

2.

"/p g

/t/lj\\.

/

o v

/

y:.

la

/

%l b

/

N..

i f

k\\.\\

. 3_

~~~.

'~s

,,, s 8'

18' 102

.O.

10 s o-s FREQUErsty-cpg FIGURE 3b s

.a o,

41 l;

)

Il 4

i.

ACCELERATION RESPONSE SPECTRA 0

4.+

..i,... r i

i

. i iii i

... ii UB101X. (RIGID DASEMAT)

UB101FX (FLE~X BASEMAT)

ABWR CONTROL BLOG

~

f4GDE 161 X'

EL 12.3M-22 DAMPlt4G i.

i.,

I-a 3.

n j

in-i o

i.

- b-i.

e 1

w

[_

gj 2.

u u

II 1

J m

a j\\^

Wu i

w g

n.

1.

tn I

W I

y i

0.

~

-1 2

3 j

10-1 10e 10 '-

.10 8

j F RE 0t1E NC Y -CPS '

j ':

.}

i-

,t

=i i

' FIGURE 4a d

[

l

.. =.

L 1.

.t b

i e

e n

2 y

0 7

1 i

i i

+

i G

O LB 7

LO RX G J

T t

4 s

t 1 1

08PP C1OMTA RE D a

W0B A

D0/2 R

AtC2 J

T 3

C 0

E7 t 1 P7 L

ST E

i l

S i

S N

P

~

i

]

O

. C n

P S

. Y R

i C

b E

J t

4

. E N

. U E

R

]

. Q U

1 i

E G

1

. R I

F i

F A

R E

L E

8 s

CT) 0

+

CrT) 1 e,

A-AT MA i

EM SE BA 4

AS i

.B D

I X i GE IL RF s

t (

i 1

X XF 1 I 0D i1 I 0B 0U 4

1 7

0 1

4 3.

p r

1 0

s n'-

os.mn ZOsF<*mJgUUC J' 4 trHgJhat t

1

~

l~

ijl 4ijjl l

I 2

0 T

1 i

e i

i i G O

. L i B L

O HZ G 4

T t

0

]

J e

t00PP C1OM TA RE D WDD A

DO/2 R

AtC2 g

i 2

l 0

C 1

i E

e P

S a

E

.S S

P i

N

.C O

P i

Y S

C a

N S

E i

E E

R U

R Q

U N

E G

I O

i

.R F

.F I

T s

A R

E L

0 E

0 Crl 1 '

C~i l )

A AT MA i

EM SE i

AS i BA B

D I X i GE I L RF

( (

Z2 Rf 1 1 00 1 1 i BB U (J

~

2 0 ~

1 4

3 2

1 O

zO. 4rw'Wuu< i_<

0j n

cm

.t I

ACCELERAT10N RESPONSE SPECTRA 4.

i i

i i i i T-i i

ie i i i i T-i 1

e e s.:

UB1DIRZ (RIGID BASEMAI)

ABWR CONTROL BLDG NODE 106 2 UB1DIF2 (FLEX DASEMA7)

EL 12.3M 22 DAMPING a>

e 3.

m tn z

O e-4 p

tr -

w d

2.

u U<

J<

E F

.u

. c w

~

l

[L 1

tn '

A v

~

' ' ' ~ '

2 0.

2 10 0

10 -

10-1 10 FREQUENCY-CPS Il' FIGURE Sb l

l4

(4)

GE needs to clarify that the UB soil profile (soll soil site) represents the lower bound soil condition afler uncertainties in soil properties are taken into consideration.

Response-The UB soil protile is the lower bound aller site uncertainty has been taken into account - This statement will be added to the SSAR Subsection 2.3.1.2.

m--

- -. - -. ~.

.-. ~... -. - -

+

(5)

'According to the statTs review experience and the studies performed by the stalT consultant, the use of Seed and Idriss 1970 soil degradation curves for the SSI analysis will lead to an unconservative result (seismic responses)c GE should con =ider to use more recent data such as the soil degradation curves recommended by Geomatrix Corporation in the analysis.

Response

j The soil profile VP3D150 was used to analyze both the reactor building and

)

control building. For these analyses, SIIAKE deconvolution analysis was o

performed using the soil degradation curves from idriss (1990), and the strain-compatible shear modulus and damping values were used in the SASSI analysis.

The analysis was performed using 3D model of the buildings in the X direction only. The results of analysis at key locations in the reactor building are compared with the respective SSI results using Seed & Idriss 1970 curves in Figures I through 3. The results for the control building are compared in Figures 4 and 5.

As shown, the new degradation curves have an insignificant efTect on the response of both structures. Based on these results, it is concluded that no modification of the results using Seed & Idriss 1970 curves will be needed J

4 1 -

o r

i.

i.

l, l

? i i )

f

!r

,,, F!

il

i-.i h

t

fl!

..s a.

• 0 i

1 i

i i

e i

i i G i

D L

i B i

E R

L OS Z

TE ZG C V X'0f 1

4 AR l

i F

EU3 P

?

RC3SM MA ClE/D lDV I

A B0OPr

,e R

ASNR2 3

T C

t_ 0 1

E i

l P

i S

i E

e S

\\

n S'

P N

,C O

i P

)

Y E-C S

V 1

U e

.N E

R

.E E

R 0

R C

0 U

) D E

G

~

N J

I O.Et i

.R' V0 F

I R0 F

T U0 C

A R

A y R

E A

1 t

E AP EP

~

L 4U 0

f 0'

J E

C.00

~

1 1'

77 C

99 i

A

. 11 e

SS i SS i

II J,

RR iDD i

II i

iDD EE EE i SS i

( (

x XA 1 I 0D

~

.33 e

~

PP

?

VV 1

~

0

,1 4

3 p

I g

ncimW ZoHW'4ELSL;.wuU< 1<%Fuw n i

l-4 l

V

T ACCELERATI0t1 RESP 0tJSE SPEFTRA 4.

i i

i i,,ii i

i i

i i i i i t

VP3D1X (SEED-lDRISS'1970 MEAN CURVE)

ABWR REACTOR BLDG

[

VP3DIAx(SEED-IDRISS 1970 UPPER BOUND CURVE)

Salt CURVES NDCE 89 X i

RCCV TOP i

21 OAMPING i

f --

m i

3.

i

.o

?

l m

Z.

.. O j

^

.m V-<m

-w A

n i

8 22 i

w y

u U<

[

g<

m H

n o

pd

(

i-w-

a 1.

V 1

m r

i e

a i

's s'i-i i

i i

n's ii i

i i

e i i i i g

8 1

2 10-1 10 10 10 FREQUENCY-CPS FIGURE 2

i l',

1I

,!1I!1,li\\:I1iIilj-,

I

!I ll!

2 e

i 1

i G D

LB i

ROS TE G

CVx t

AR i

EU5PP RC9OM TA rte D wlDB A

B0U/2 R

ASNR2 T

[

t C

0 E

~

1' P

S r~

E S

.ii-

\\

S N

P O

C P

)

E p,'

Y S

V ij R.

R C

E U

N

'3

- C

.E

' E c.

U N

) D Q

RU O.EN E

G VU I

RO v

R I.

T UB F

F C

A R

R NE E

AP EP L

MU E

8 00 Cr77 0

1 C

99 A

1 1 SS

.SS II RR

.DD II

.DD EE u.

EE SS

( t X

xA 1 1 D0

.33 PP VV 1.

0_

1 6

4 2

'O' a4 mn ZomF<mo3aUu< J<msottn otn t

t t

m L

i- <

t i.i'

,i 2

ci 44 ' '.i j

< < j

.i

y 1

l ACCELERATION RESPONSE SPECTRA 1

4.

'VP3 Dix (SEED-IDRISS 1970 MEAT 4' CURVE)

VP3DIAX(SEED-IDRISS 1970 UPPER BOUND CURVE)

ABWR CONTROL BLDG NODE 161 X EL 12.3M 22 DAMPING i

i cn.

i 3.

.m

, tn 2

o m

b 1

4

.E tL2

.._J p.

g u

t-u 4

J m

V

.i Uw 4

la g

g

\\

I

\\u o.

8 2

10-3

'10e.

10 -

10 il FREQUEtiCY-CPS

~ FIGURE 4 p

!~

.a

ACCELERATION RESPONSE SPECIRA 4

.... r -

.. i i.-

VP3 Dix,.iSEED-IDRISS 1970 MEAN CURVE)

- - VP3D1AX(SEED-IDRISS 1970 UPPER 800f4D CURVE)

ABWR CONTROL BLOG N00E 181 X C/B TOP 22 DAMPIr4G 4

cn 3:

<a tn z

O

- M F

r

<m W

_a p

W-i o

u

=

.m

- F o

q W

AA 1

E VV 4

0.

4:

10 ~ 3 10 10

- 102 8

2 FREQUEt1CY-CPS ~

4 FIGURE 5 i

4 4

i 4

em.I-[,

,s..

n i-m

,-.y%u,c-,,

+,-w 5 y

y.,,

.-r-3.-

y,

,_,,,, ws,s...

w a',,_..

d (6)

Gli should provide the bads and revise the SSAR to state the failure of the reactor building stack will not hase any safety impact to the seismic category I structures.

Response

The reactos building stack shall be designed to withstand the effects of an SSE. It shall use a dynamic analysis as documented in 3.7.3 for subsystems. The response spectra for the building c!cvation that supports the stack.is provided in Appendix 3 A. The height of the RB stack will be determined on a site specific bases to help utilities meet 10CFR20 and 10CFR50 Appendix 1. A COL action item will be added for the stack t

b

. 1 I

h 1

r m

4,.

g...

-., ~

+--h-na a

,r-e,w

~

e a-n

,m

,,.~._...,

..._..,y

... ~. _ _ _ _ _ _ _

- ~ _ ___ _. _ _.

r; i

,4 x

(7)

GE should confirm that the soil pressure used in the design of embedded building walls is adequate in comparison with the soil pressure obtained from the SASSI

~

analysis including structure to structure interaction en'ects ltesponse:

Soil columns were modeled on both sides of the reactor and control buildings in the respective SASSI 2D models. Both buildings were analyzed in the horizontal X direction using three soil profiles UBID150, VP3D150 and VP5D150, representing son soil to son rock profiles. The seismic soil pressure on the walls -

was obtained from the SASSI analysis In addition, three 2D structure to-structure cases including reactor, control and turbine buildings ni deled in one SSI model were analyzed using the abose soil profiles and maintaining the soil columns _-

on each side of each building model. The seismic soil pressure on the walls of each building were obtained from the envelope of seismic soil pressure nom the six SSI cases. The results on two walls of each building were subsequently enveloped to.

i obtain one seismic sail pressure pro 0le for each building. The enveloping results for the reactor building and control building are shown in Tables I and 2 respectively. The seismic soil pressure results shown are conservative since the effect of structure inertia are considered in computing these pressures.

The exterior wall of the reactor building that faces the control building and the exterior wall of the control building that faces the reactor building have been l

evaluated for the soil pressures shown in Tables I and 2 respectively. For the reactor building wall it is found that the concrete and the rebar stresses are within the allowable limits Ilowever, the shear ties were over stressed by approximately 5% and therefore additional shear ties of #8 @ 300mm x 300mm have been l

provided for this wall from TMSI 8200 to TMSL 12300. _ For the control building wall it is found that the steel area provided in the design is adequate.

3 b

w l-V 1-1 l

m

@r"K'TWW TrwW wge ry DWmwT'N...

.. w wywmW w w,

..WW*=wy'363 y ' rN ww1'gu.r=W W wrr.y 'y F r yw v my-W ww-'3*h*W"W"Y TwryW mv

-Y 9

4 TF'

l l

TABLE 1 SEISMIC SOIL PRESSURES (Reactor Building)

ELEVAT)ON MAXIMUM VALUE 34'l(m)3ii?" @O {t8 pip? lW 12.00 94 9.90 9.90 44 7.90 7.90 22 6.30 6.30 16 4.80 4.80 15 3.50 3.50 12 0.90 0.90 17

-1.70

-1.70 24

-3.80

-3.80 16

-5.90

-5.90 40

-8.20

.., ~ - - -

TABLE 2 i

SEtSMIC SOIL PRESSURES (Control Building)

. ELEVATION i MA10 MUM VALUE g jyy;F" @' f(tshiUv '

12.00-94 9.90 9.90 44 7.90

{

7.90 26

. i L

6.30

-6.30 14 4.80 4.80 22 3.50 3.50 14 0.90 0.90 17

-1.70

-1.70 24

-3.80

-3.80 34 1

-5.90

-5.90 40-

-8.20

?

t f,---,-

--,,.u

.-y, y

.v,

.---.yw.c 3. -v p

(8)

GE needs to provide the tesis and the documentation for the devHopment of the enriched design ground response spectra

Response

GE provided this to the statTin an earlier transmittal 9

49

.___-_.-_.._-______----______..--____.--,_-_---___.i______-_---______

w

i i

(9)

GE should update the SSAR, design drawings and ITAAC information to be consistent with the tier 1 information

Response

iI The drawings and dimensions will be updated when the SSAli is updated afNr the second audit. Some markups for the SSAR portion have already been updated and-are included in the markups provided.

4 1

1

1 (10)

GE should provide the Japanese results associated with the distribution of seismic -

forces and moments between various sticks of the model and compare the results with GE's model a

t

Response

t

+

GE is still negotiating with our Japanese associates to allow us to provide them to the NRC. GB will try to provide these in a later transmittal.

i E

-.[

I r

9 I

1

-k

,-.;,-,i'-~..-,.,-,.e-,'.---.,.-,

---

..,,,,,,,--,,,,.w.,,

.:nn-..

n

...wa.

(l1)

Two major concerns were raised regarding the Ril seismic model:

(a)

GE should recheck the adequacy of the assumption of double synunctry for the Ril.

Response

An eigensolution and a time history analysis was run using the 3D finite element model developed for the RIl stress analysis and the SASSI stick model using STARDYNE Table 1 compares the fundamental frequency in each of the three directions of both the 3D FEM model and the SASSI stick model for the fixed base condition. As can be seen, the two models predict essentially the same frequency. Using the sarne acceleration time history as was used for the SASSI runs, a time history response run was made, and response spectra generated. Figmes I through 5 compare spectra for the Ldirection runs. Figures 6 through 10 compare the Y-direction runs, and Figures 11 through 15 compare the vertical direction runs. These comparison indicate that fbr the horizontal seismic motion the results of the stick model match well with that of the FEht. For the vertical ditection the results between the stick model and the FEM show a good comparison for the RCCV (Figure 15). For the Reactor building walls the spectra in vertical direction (Figure 11 through 14) show good comparison up to about 10 lh But in the frequency of about 20 liz, there appears to be a frequency shift This is attributable to local modes of the FEM model llased upon the study results of modal analysis discussed above, the ftmdamental frequencies are matched within 1% in all three directions llased on the results of the time history analysis discussed above, the ARS curves are also matched satisfactorily. From these discussions it is concluded that the stick model and the FEM are dynamically equivalent, and the assumption of double symmetry for the RIl seismic model is adequate.

(b)

The tiexibility of the foundation mat needs to be considered.

Response-See response to item 3.

~-.

Table 1 Comparison of Modal Properties Model X Direction Y Direction Z Direction SASSI Stick Model 4.14 3.92 9.53 3D Finite Element 4.15 3.90 9.63 Note: 1 Stick Model represents full structure 2 Finite element model represents half structure

2

'0 1

i i

i 7

-r i.

7

,1 3

7 7

L E

G SG N

~-

LN I

R Ll P

Ax M

W WA A

B H

D A

BS

/ -

2 8

Rx 2

0 1

A i

R i

T i

C E

e P

Z S

[

H i

E Y

S C

i N

N E

O U

i P

Q S

E R

~

E F

R i

N 0

1 1

A se R

f' l

E-T L

i E

C iK C

E A

IE i

T Sf i

i x

/

XX 2

x1 e

83 32 ru EE iDD g

OO i

NN F

t 0

0 B

S 4

2 91 1

n 31)ao)L4 #4EVu) tot

~

ma mn 2oset<%t a

n t

t

.,1 L

4

0 1

m 35 3

2 7 L L

G SG N

,LN I

LI P

AK M

R VI A

D W

P B

/ -

1 A

RX 2

8 0

1 A

R T

^-

^

C E

P Z

S H

E 3

Y S

C N-7 N

E O -

U P

Q S

E R

E F

R N

0

./

1 1

A 8

R 0

1 u

r c

,K c,C A

l E i

Sf X

X -

n

- x X9 2

,07 03 12 e,EE r

DD u

OO g NN i

F 8

0 0

B G

4 2

, 1 1

2_4&tuwom n

vtgw33uu4 oamf ZOe l

- 1

(

(

a

I fl, lfi il!il' Il,

\\l\\l)!;!l1 j\\!\\l' 20 1

,3 2

1 t

E G

SG N

LN I

,LI P

AK M

R WA A

H D

4L 8S O

/-

2 Rx 2

A

'0

\\

1 A

R I

C E

P

~

Z S

H

, i' A {.

Y E

f S

CN N

E O

U P

Q S

E E

R R

F N

O I

I A

e 0

l R

1 E

L E

C,K C,E A

IE T

,SF i

X

,x-3

- x x5 31 05 e

11 ru EE D0 i

4 O0 NN F

,0 0

B 6

4 p

1 1

as~m ZO~t<xWJwuO< a4mFuWG

!fl f'fif[llril!f1l!lIIl l!i!I1lIIIl j

tI I

Ill1ll!l 1i

I l

Ii I

20 1

.r 3

~

8 4

L

(

G SG N

L N I

.LI P

AK M

~

R WA A

H D

4 k

BS B

/ -

2 Rx 2

A

'0 1

A

=

R I

C

~

E P

Z S

H E

Y S

C 4

N r

E 0

y U

P Q

S E

E R

R F

4 tu l

I A

8 R

0 1

E t

E C

.K C

C

.l E

A i

SF X

't x-

- x m

x2 44 e

88 r

11 u

9.EE DD R

DO HN 8

-,6 0

B 6

4 p

1 1

c en zome4mwswuo4 J.4EnUu(1n s

n t

a

F.gure 5

ACCELERAi10N RESP 0t1SE SPECIRA--AUWR

,_,_,_,7.

16 i,,

t-RCCW EL 23 S.

t40DE SBx-x SilCK x-SHAAING NODE 2.49X-X F.E.

22 DAMPING 8

m e

t tn Z

O C' 6

'-):;

4 tr ta

_a to u

o 4

4 2

4 tr k

i I

F o

to g

(D 2

2 8

10 8

10 10 10-1 FREQUENCY-HZ

'0 1

~

.7 3

3 E

,sG tN I

tI P

A K M

R WA A

W H

D B

85

/ -

2 A

RY 2

A

~.

'0 1

R T

C E

\\

P i.i Z

S

.:i:i H

i'.

i

'.i'

i E

Y

.A' S

C N

N

N E

O U

P Q

S E

E R

R F

NO I

T A

8 R

0 1

E L

E 1 C

C M

A.CI E T.

Sf Y

YY 6

T6 e

B7 r

S2 u.EE g

D0 i

O0 F

NN 2

l -

0 0

B 6

4 2

61 1

e*=n zoer<Ewa_'uU< s_4gnuj.n n

tQt a

i n

l

R ure 7

ACCELERAT 10tJ RESP 0tJSE SPECIRA-- ABWR rr,,

,7 7 3

to R/B Matt EL 23. 5..

N00E 10BY-Y STICK Y-SHAKING t< ODE 2233Y-Y F.E.

22 DAMPING 8

m 8

an

l Z

O 6

4 E

ta E

to Uu 4

s 4

E s-

.i U

td a

tn 2

10 3g:

0-las 10-8 FREQUENCY-HZ l

.)

.o Figure 8

ACCELERATION RESPONSE SPECTRA--ABWR j

10 R/8 WALLS EL 12.3=

NODE 193Y-Y STIEK

)

Y-SHAK]NG

- NDDE 1375Y-Y F.E.

22 OAMPING I

B on I

=

en zo we 6 4mw 3

w u.

- o

-4

._)

\\

<t m

ii e

u i-u) 7 i

a e

n t's 2

i

\\

' '.' 0 O

8 1

e

- 10' '

j e FREQUENCY-H2 F

0 1

.e 4

L E

G

.SG N

LN I

R L I P

AK M

-~

W WA A

~

D H

D A

0S

/ -

2 RY 2

8 0

A 1

R T

C E

P S

Z H

E Y

S C

N N

O E

P

.L U

S Q

E E

R R

F N

0 1

l A

6 R

0 E

1 l.

E C

.K C

C A

.I E

T Sf Y

9 Y -

- Y m

YS 40 09 e

1 ru EE DD a

i OO NN F

8

-,0 0

0 G

4 2

1 1

m,om zOMr<EwJwuo< J<Ttuwn*

s lllfIIjI;lIllliIlliIl>i\\!!l!

l!1!

O

,g

~.

5 3

G 2G N

N I

l1 P

R Ex M

A A

W VH D

B CS A

C -

2 RY 2

g G

3 A

4-R C

.v T

4 E

PS j

Z I

I I

H E

Y j

S C

N ir O

E P

U S

Q E

E R

R F

N 0

1 I

A s

R 0

E 3

L E

C E

K A

,C IE T

SF 0

Y 1

YY m

3 Y4 e

B1 r

S2 us,EE e

DD i

OO F

NN 8

-0 0

B 6

4 2

e1 1

2<EHuwQ ncien 2OmE4mw>w"u n

v3 F gure si ACCEL ERAl lutJ RESP 0tJSE SPEC TRA-- ABWR 6

R/8 WALLS EL 31.7.

NODE SBZ-Z SilEK NODE 23122-Z F.E-I-SHAKING 22 DAMPING m

a e

tn Z 4 o

s t-<

Tw

_J w

u W

4 2

_J 4

W 2 F-

~:

O

^e I r;

in c.

39 10e ist 0

10 FREQUENCY-HZ

Fi ore 12.

ACCELERAIION RESPONSE SPECTRA--ABWR S

6 NODE 100Z-2 STICK R/8 MALLS EL 23 5 N00E 23753Z-Z F.E.

z-saaxtug 22 DAMPING e

e tn

. z 4 i

. O s

V-

-4 to) tshu u

4 ti J

<c.

^

i W 2 t-L)

.--~.......;

i,

f Ed 1

O.

-m

-j

\\

r.-

o 10 10 108 102 8

FREQUENCY-HZ am J

w m

m.--

Figure 13 ACCELERATION RESPONSE SPECTRA--ABWR 6

NODE 1932-2 STIEK R/B WALLS EL 12.3.

- - - N00E'15152-Z F.E.

Z-SHAKING 22 DAMPING on 8

m tn z 4 o

e-.

E<m w

2 wuo 4

_.J<

E 2

.4 i

s--

U

• i tij

! i

^k l '

i 1

g

-f i

\\

r.

10-1 10 108 102 8

FREQUENCY-HZ

Figure IT ACCELERATION RESPONSE SPECTRA--ABWR t

s N00E 1042-2 STICK R/8 NALL S E L

4. 8.

- NODE 1842Z-Z F.E.

z-Saaggug 21 DAMPING cn i

.IB m

z 4 O

w r-<

K tu

-.s tsh ou a

E 2 s

U

\\.

M M

r.

'10'8 10e 10' 102 FREQUENCY-HZ

Figure tY ACCELERAI1ON RESPONSE SPECTRA--ABWR e

NODE 902-Z STICK RCCV El 23.5 NODE 2149Z-Z F.E.

7_Sgggggg 22 DAMPING m

i 6-e tn z

A o

e-o F

<mw

' 4 w

U U

_s E

4 e

W U

w O.

2 y-tn

.r-g V

10-8 10s 101 102 0

FREQUENCY-HZ

I (12)

GE should assess the need to combine the cracked RB model with a site condition that has a fundamental soil column frequency around 4 IIz in the re analysis.

Response

The soil case of VP3D85 with control point defined at 85 fi depth (convolution case) has a soil column frequency close to 4 Hz. his case (VP3D2CX) was-analyzed using 3D model of the reactor building in X-and Y-directions, The results are compared with the respective soil case using uncracked properties (VP3D2X) as well as the upper bound soil RlU case (rigid soil, uncracked) and RlC case (rigid soi, cracked) results at the key locationdn the reactor building in figures 1 through 3, As expected, the response amplified at about the soil column frequency.

However, the results of this case will not be included to define site-envelope loads since the convolution case is not a design condition for the standard plant.

ACCEL,ERATION RESPONSE SPECiRA 10.

i t-i i

nT i

i i

i e i1 ;

i i

TrT i i i RI UX (Ut4CR ACKE D)

ADWR RE ACTOR BL DG.

RICX(CRACKED) 00tJCRE TE CRACKING.

-- VP3D2X(UNCRACKED. Cot 4V) tJ000 33 x

- - VP302CY (CRACKED.. C0tiv 3 RPV/MS F40ZZLE 22 DAMPlflG 8.

cn i

rc m

Z o

(

sr 6.

i

,l e-a d

Ild

~,

W l.1 u

i!g:\\

u<

l

/ j.. ' -) :\\

i 4'

1, li

__s

/r As.-

\\\\\\

f e

v

./.

U

\\

\\r \\-

w

/

,f y f

l a.

2-

_,!a I

\\.1

_\\

hr l'\\

• ^~.- i

' /f f M

\\.~

v 0

\\,i s._

0 10-8 10 10 102 8

1

. FREQUENCY-CPS FIGURE la

3--

8 4

ACCELERATION RESnONSE SPECTRA 10.

i i

i i..i i i

i iimii R1UY(UNCRACKED)

ABWR REACTOR BLDG.

R1CY(CRACKED)

CDNCRETE CRACKING.

--- VP302Y (UNCRACKED. CONV)

HDDE 33 Y

- - VP302CY(CRACKED. CONV)

RPV/MS NDZZLE 2I DAMPING 8..

m-I' m.

W Z.

l(j o-m F

6.

l l }l m

W w

l a

o l

lj\\

s u

(

4 I

', eg I.I 4 Il 5

ly!\\

l.$ t. k m.

f l

I I.

lf'\\j.y {\\

u-f w

c.:

\\.

)

a..

l :'

i?. kn

\\ l

.;I

d jp

~ ~ \\ ft)Iy e

)lf 6

M ' /,'.

\\.

0.

-~

10'8-18e 10 102 2

FREQUENCY-CPS FIGURE lb

ACCELERAl10N RESPONSE SPLCTRA 10.

i i

iiiiii i

i i

i iiTi i

i i

i i e a i R1HX(UNCRACKED)

ABWR REACTOR BLDG.

RICX(CRACKED)

CONCRETE CRACKING.

--- VP302X (UNCRACKED. CONV)

N000 89 X

- - VP302CX(CRACKED, CONV)

HCCV TOP i

22 DAMPlHG

'5 j

Il.

8.

lg.

a I

{;.

2 iM

lp z

i o

.i t.

F 6.

Jy i

1 e

I I

t. \\

wa I - l \\.1 w

o I, y. 1:

N o.

k 4-jj.y

}\\..

1 I ll..\\1 1r:.

.m Ya.

tP

\\\\\\

8

!"Il M.\\,

~i. * )..

fr j 'u 2.

% '\\.

N' i

0.-.

e 2

10-1

'le.

1g 10 FRE QUEtJCY-CPS -

FIGURE 2a

12 ACCELERATION' RESPONSE SPECIRA 1

T

-i iiiii i

i i

iir2 i i

i i

i ri

,a R10Y(UNCRACKED)

.R1CY(CRACKED)

ABWR REACTOR BLDG.

-- VP302Y(UNCRACKED. CONV)

CONCRETE CRACKING.

'- VP302CY(CRACKED. CONV)

NODE 89 Y RCCV TOP 22 DAMPING-g 11 f

9.

a*

i I

z I

o l

i l :bp.

W

-m' l ~ {11 i

to J

6~

ti

['

/N N

/5*;

I }Al.L.\\

s 4

\\

x l

H l

l

\\.

La 3.

"I$

h\\ \\

m

/ o/

j\\

/,/

\\.h -

a/

i.4

,A[*'

\\

M j

- wS w-0.

'J-

'10'2 1Os 2

10 -

'102 FREQUENCY-CPS FIGURE 2b

ACCELERATlON RESPONSE. SPECTRA 16.

i i

i

, i i ii i

i i,iii i i,,,

R1GX(UNCRACKED)

ADWR REACTOR BLOG.

-RICX(CRACKED)

CONCRETE CRACKIfiG.

--- VP3D2X(UNCRACKED. CONV)

NODE 95 X

-- VP302CX(CRACKED. CONV)

R/B TOP 22 DAMP]NC E

cy) i 12 so

i

)th m-I'i z

$ ::ll o

N

.! li H

l

• gi tr tij I4.

i,

. I ::j..x.I ;

u p.

5 1

4 l

l [*::

a:

7. 'l V:. ;.

H o

B.:

Ty\\

i g

4*

N i

'd ~.

(

s ;.j '

nn

.n\\-

\\/

1.MV(E

/

gf/q-m Q.

==

.=.x j

\\---_.____

0;

' 3 g3-102

'10-2 10e FREQUEfJCY-CPS FIGURE 3a.

9

~ _..

.m

-.w g

I ACCEi. ERAT I ON RESPONSE SPEC T RA r rTri r

i i

16.

i i i ir i

i

.. i1 i R10Y ( UrJCRACKE D)

ABWR REACTOR BLOG.

Cor4 CRETE CRACKING.

- RICY(CRACKED)

NDDE 95 Y

- - VP3D2Y (UNCRACKED. C0tJV) 0

- - VP3D2CY (CRACKED. CONV >

vl glir cn pi i

12.

l m

i i

in

! 'I:

!I Z

i

!. I:

O Si t-g 15 I'

h

_a 0-f ;i

]

w g

O l:: )

o I*I Y

1 :-l\\.

b,.

8 l

4

\\\\

x

.h e

I l.l uw j

n

,_ 3 l..i tj v 4-f i/

.h 1:

ll'

~j

- Qj y\\

\\

Mpi A,h x.==._._..._-

v n

v

- c' 10 ~ 2 10e-3gi 302 O.

FREQUEflCY-CPS FIGURE 3b i

y

3 (13)

The stalTraised several concerns associated with the floor respons, spectra from the rigid soil model of the RB:

(a) l The cause of one single spectral peak from the new model and two spectral peaks from the old model.

Response

The cause for the difTerences in spectral peaks is attributed to the structural model update. However, the ditTerences are not so pronounced when.

comparing the results for the same soil case As explained in the response.

to part (b) of this comment, a new soil case (VP7) was added to the re analysis. -The resulting response of case VP7DSX at the top (node 95)is given in Figure 3a, which shows two peaks. This is qualitatively similar to the original result (TIOX) given in Figure 3G.2-20 of the SSAR~

(amendment 4), although the second peak of the original result is more pronounced than that predicted from the updated model This is attributed to the. super-structure modification for tornado resistance. As to the elTect of model changes on the RCCV response, a comparison is made at node 89 ~

between the new case VP7D5X in Figure 2a and the old case T10X in-Figure 3G.2-20 of the SSAR. A single peak response is essentially-observed for both results. The frequency response characteristics of the RPV at the MS nozzle location (node 33) are noticeably different (see -

figure la of this response and figure 3G.2-20 of the SSAR), due to the elimination of the RSW stabilizer from the current model.:

(b)

GE to justify that the structural response from the rigid soil condition bounds the responses from other site conditions such as the site with a shear wave velocity equal to 5000 fVsec.

. Response:

To response to part (b) of this comment, a new case was analyzed for a.

rock profile with Vs=5000 fps. The results of the analysis for this new case i

(profile VP7) are compared with the upper bound soil profile R2U (rigid -

soil uncracked with no separation) atlthe key locations in the reactor building in Figures I through 3 in X,.Y., and Z-directionsD As shown, the results of VP7 case are generally enveloped by the results of the upper l-bound case. A relatively small shift in_the peak frequency is observed as L

expected.- The VP7 case, however, will be included as one of the'SSI cases -

for both the reactor and control buildings.

l

-~

(13)

The statTraised several concerns associated with the tloor response spectra from the rigid soil model of the RB:

(a)

The cause of one single spectral peak from the new model and two spectral peaks from the old model.

Response

The cause for the dilTerences in spectral peaks is attributed to the structural model update. However, the ditTerences are not so pronounced when comparing the results for the same soil case. As explained in the response to part (b) of this comment, a new soil case (VP7) was added to the re analysis The resulting response of case VP7D5X at the top (node 95)is given in Figure 3a, which shows two peaks. This is qualitatively similar to the original result (TIOX) given in Figure 3G.2 20 of the SS AR (amendment 4), although the second peak of the original result is more pronounced than that predicted from the updated model This is attributed to the super-structure modification for tornado resistance. As to the elrect of model changes on the RCCV response, a comparison is made at node 89 between the new case VP7D5X in Figure 2a and the old case T10X in Figure 3G.2 20 of the SS AR. A single peak response is essentially observed for both results. The frequency response characteristics of the-RPV at the MS nozzle location (node 33) are noticeably dilTerent (see tigure la of this response and figure 3G.2-20 of the SS AR), due to the elimination of the RSW stabilizer from the current model (b)

GE to justify that the structural response from the rigid soil condition Sounds the responses from other site conditions such as the site with a chear wave velocity equal to 5000 tVsec.

Response

To response to part (b) of this comment, a new case was analyzed for a rock protile with Vs=5000 fps The results of the ancivsis for this new case -

(protile VP7) are compared with the upper bound soil profile R2U (rigid soil uncracked with no separation) at the key locations in the reactor building in Figures I through 3 in X, Y, and Z-directions. As shown, the results of VP7 case are generally enveloped by the results of the upper bound case. A relatively small shit) in the peak frequency is observed as expected. The VP7 case, however, will be included as one of the SSI cases for both the reactor and control buildings.

(c)

The vertical spectral peak from the re analysis is substantially higher than-the original vertical spectral peak which is equal to 2 times the OBE spectral peak; Justification is needed.

Response; The observation made in part (.c) of comment Noel 3 is related to the comparison of the vertical response at node.89 between the rigid soil case (R1U) of the re analysis and the hard rock case (HRD85) of the original-analysis. In addition to the model difTerences as stated in the response to part (a) of this comment, the use of different site conditions also attributes to the differences of the two results. In the response to part (b)'of this comment, a rock profile of Vs=5000 fps corresponding to HRD85 of the original analysis is added to the re analysise The resulting 2% damped -

response spectrum at node 89 in the vertical direction of this 'new ~ case -

(VP7D5Z)is shown in figure 2c The maximum spectral peak is about.

3.4g, which is close to 2 times the OBE spectral peak if the original analy"s case T10V as shown in figure 3G.2 21 of the SSAR; amendment

4. Furthermore, the ditTerences can be resulted from using a new input-time history that better fits RG 1.60 spectra.

(d)

The vertical spectral peaks at nodal points 95 and 89 occur at substantially different frequencies. An explanation is needed.

Response

It should be noted that in the upper bound SASSI case (RIUZ) and the STARDYNE analysis, the nodes corresponding to the reactor building-walls in the ground are confined to the rigid soil medium. As such, the free -

standing height of the reactor building walls is reduced resulting in higher stifTness and higher frequency of vibration, as shown in figure 3c of-response to Comment No, I for node 95.?The RCCV, on_ the other hand, is not directly constrained by the rigid soil medium ~ lts free standing height--

is longer, thus, resulting in lower frequency of vibration as shown in figure -

2c of response to Comment No. l for node 89.

~

L 2

^. l

^

b A

,.)

'e' ACCELERAT10N RESPONSE' SPECTRA i

3

• i3 i

i i e iii R20X i

- VP7DSX ADWR REACTOR BLOG.

NODE 33 x-RPV/MS N0ZZLE

,.- o 22 DAMPItJG cn i.

6.

sn zo s

F 4

E LU

.J 4-I La Uu

_3 2

Ee

3 LJ a:

ta a-2.

3 0.'

t.

i i

i

-.iL 10-2

3 ge

- FREQUENCY-CPS

- 10' 2

10 -

' FIGURE Io..

t il f

v g

g j'

-[*

v.

d g'

-.r w

v

%.O' m

'u y

g

a v,,.

10...

ACCELERATION RESPON5E SPECTRA i

i i

i i i ii a

i i

i iiii i

i e

i e i 4>

ABWR REACTOR BLDG.

R20Y NODE 33 Y VP705Y RPV/MS N0ZZLE 22 DAMPlNG 6.

cn -

I-to tn Z.

O e-e.

i F

6.

.x tu J

tuu F.i o

':]

4.

i

' \\

4

.'9 m

F U-

-- td -

~-

R m

. ')

0 2

710 101 50-5' 10e -

FREQUENCY" CPS

' FIGURE-/6 j

ACCELERATION RESPONSE SPECTRA 4."-

i i i i ii i.

i e,,

i i

i i

i c, i i R20Z ABWR REACTOR ULOG.

VP7D52 t4DDE 33 Z-RPV/MS f40ZZLE 22 DAMPItJG

cn

's 3.

-m.

(n.

Zo

.s V-<

.Ew

-J p~

\\

w

'o

-i o<'

~.

.g.

m-

.,. : f.

i p.

U

~

W J

n..

1, tn

. 0 ;

10e.

10 '

'102 1

10-1 FREQUENCY-CPS

.c i

FIGURE /c

s.,-

,1 M

Y 20 i

1 9

i i

i s

i G O

L i B RO T

G J

CX t

i A P]

E9OP R8T M A

REVD WDC A

D0C2 4

R At R2 5-T 0

C E-i 1

P T S

i

,:1 j

E e

.~

S S

i P

N i

C O

Y u

P i

S C

Z 1

E a

E E

t R

U R

Q U

N

- E G

I O

i R

F I

.F T

A R

E L

.e E

0 Cr 1

C i

A i

i i

i e

X 5

XD 07 i 2P RV 3

T 0

1 0

8 6

4 --

p 0'

1 n: an ZOsFtgwJWo'ot JtmFUwCtLD--

cI s n

)

4 i

2 0

1 i

i i

i i

i G D

L i B RO T

G 4

CY t

i A Pi E9OP R8T M A

REVD WDC

\\

A B0C2 h

R At R2 t

T g

C E

i 3

P i

S i

E i

3-S S

. P i

N C

O S

. W 4

P

~.

Y 6

C 2

N R

i

? g:

E E

E U

R Q

U N

_ E G

I O

R F

F I

T A

R E

L 8

E 0

1 C

C i

A i

i i

i i

Y S

'Y D i07 2P RV 2.

2.

0

,1 0

B 6

4 E

0~

1 n am zost<Tw1woU< J<xtuW1m cI

(

1

-ACCELERATION RESPONSE SPECTRA 0,

i i

i i, iii i

i i

i..ii i

i i

i. i.... >

'R20Z ABWR REACTOR BLDG.

i

'VP705Z tJ0DE 83 Z RCCV TOP 22 DAMPir4G

. cn i

6.

ta

.. (f)

Z

-l O

1 s

\\

V-<

1 x

tu l

~J 4-to i

o

'U<

a<

m W

U.

f ta 0-2 j-

~. _

p-

' ' ' '. '10

~ '

0 2-

. j 0-1 10 -

.101 0

' FREQUENCY-CPS FIGURE ; 2c.-

I 1

~

v

N e

d,.

fI

+

r' 4

4

'7

~

i.

i y

e 2

w 0

'l i

i

+

r-

'=

=

,G e

.D L

iO '

4

=

R"

^

4 O

~

T G

i CX t

iA i

E5PP R9OMTA

+

RE D

WD0 A

B0/2 R

At R2 A

2 i

2 T

l 0

C 21 E

. ~. -

P i

S i

i i

E S

i S

P Y

N

.F C

O I

P i

7*

Y S

~A C

3 l

J

. K-

'E E

t R

i EU P.

-e D

il G

N E

I O

e i

R F

I F

g

+

r T

A f

R e

E L

8 E

0

/..

1 C

i 1

C i

r A

-e i

i m

i i

i s

X

/

X0 i07 2P RV i

r 2.

w 7

0 1

4 M

1

- 2 0

8-6 4

y

.g 4

^

1 1

y n

d ;_ a

. ut 2O h<ei. LOU < s_4KgotcW eij

~

~

' I

F; ACCELERATION RESPOf)SE SPECTRA 14 i

i i

>iir i

i r

R20Y-ABWR REACTOR DLDG.

VP705Y FJ0DE SS Y R/D TOP 22 DAMPir4G '

12.

.m b

d 10.

j Zo-l-

F 0.

e:

w g

W i

u u

6:

.3 l

m F

'4 1

-u 4-w n.

m 2.

l

. \\..

p

[

'~

~

2

~

.g.

i 8

2 10'1' 10 -

108 10 ~

FREQUEtiCY-CPS i

-FIGURE Sb i

c s

ACCELERATION RESPONSE SPECTRA 6.

i i

i...

i i

i i

i i

i i R202-ABWR REACTOR BLDG.

- - VP705Z NODE 95 2 RCCV TOP 22 DAMP]NG cn '

' I m '-

a z'

4.

O-w H<

m W

J

. w

}

u i

y

[

U 4

/

J

>:5 h.l-i n.

.tn i"'

-0:.

10.8

-10 107 2

10-2 FREQUENCY-CPS-f FIGURE 3c.~

~

,y

-(14)- In order to adequate 1t calculate the soll pressure ibr theasign of the embedddd.

walls, GE should add finite element soil colunes next to RB and between RB and L CD in the structure-to stmeture interaction model

Response

See the response to Comment No. 7.

l

_ _.. ~

(15). GE should compaie the results obtained

. (a) 2 D RB and CD individual cases and i

i (b)

-2 D s:ructure-to structure nteract on case

Response

- 1 Both the reactor and control building were analyzed individually for three soil profiles UBID50, VP3D150 and VP5D150 using the 2D _model of each building in X-direction. The analysis was repeated using the multiple 2D models of reactor,,

control, and turbine buildings in one SSI model to consider structure-to-structure iveraction etTect using the same three soil profiles. ' The results of analysis for the reavor building are_ compared with the respective 3D resuits and the upper bound case resultiin the horizontal X-direction in figures 1,2 and 3 for soil profiles -

UBIDi SO, VP3D150, and VP5D150 respectively.

As shown in these figures, the effect of structure-to structure interaction on the reactor building response is relatively insignificant _ and in general, tends to reduce the overall response is relatively insignificant and in general,~ tends to reduce the overall response of the building, The results of structure-to-structure interaction.

cases are either enveloped by the respective 3D case results or the _.esults of the upper bound case.

The results for the control building for the same three soil profiles are shown in -

figures 4, 5, and 6. As shown in these figures, the effect of structure-to-structure interaction is more pronounced in the response of the control building. The SSI

~

frequency of the reactor building is present in the control building response through a dip in the response spectru_m. Similarly, t_he results of structure-tu.-

structure interaction cases are generally either enveloped by the respective 3D case results er the results of the upper bound case Based on these results, structure-to-structure interaction effects will not be -

cons dered for the purpose ofobtaining seismic response of each building. These 1

efTects, however, will be considered for computing seismic soil pressures (see response to comment 14).

l-l-

l-4 e

4

--*-54e,--

t-r-

- 4 w-re e

s-

-e e-,

..+,..,e me e+

-r e'--e w

(

1 ACCELERATION RESPONSE SPECTRA

.l 8.

i i....i i

i i

.. is i i

i.

R20X-

--3D R/0 ARWR REACTOR BLOG.

UB10150 - 3D R/B STRUCTURE-TO-STRUCTURE NODE 33 x

- 0810150"'- 2D R/B RPV#fS NDZZLE

-- U010150 - 20 R/B+C/B+T/D 22 DAMPitiG

.' I

\\

i v

a i

G.

1

. ro m

J z

o l

e-+

F-

.4 u;

A*

ui

'U U<

A E

.x H

U uJ Wi Qp\\

..?

S\\.h5

[. ',

\\A/;

' D. \\

/ g]\\

^\\

1.

~

\\

... _(

g LAS o 0.

2

10 '

10 2

.10-1

~10e

~

FREQUENCY-CPS.

FIGURE la~

....... ~.. - -.. _..... - _..,,

' ACCELERATION RESPONSE SPECTRA B.

.... ii

.R20X

- 3D R/B ABWR REACTOR BLDG.

- 'U810150 - 3D R/B STRUCTURE-TO-STRUCTURE

- - U810150 2D R/B NODE 89 X

- - UB10150 - 20 R/B+C/B+T/B RCCV TOP 22 DAMPlf1G m

6.

m

.tn JZ.

-.. o.

m

. ta.

la' 4-u U<

J<

'T H

U tu (L.

2.

li ^

r e., P.

t f' - )\\.(<._\\.

q f-i v

j,

. Af.j

' D.\\ -

,(-

i

.Q J

s-f

\\ * \\.&%

4s a C W _.= =_=r..==.. =_ r

~0.

~

' ' l' ' '

1 2

10'1

.108

.10

-10 FREQUENCY-CPS J

- FIGURE lb m

m i.

m

.m w

=s

.r w

-.A

^

ACCELERATION RESPONSE SPECTRA

,,iii 10.

i i

i

' ' 3 -

ADWR REACTOR BLOG.

- - R20X

- 3D R/B STRUCTURE-TO-STRUCTURE UB10150 - 3D R/8 t40DE 95 X

- 0010150 - 2D R/B R/B TOP

--- UB10150 - 2D R/B+C/8+ T /B 22 DAMPlt4G 6.-

m Ia on zo s-Ar 6.

<ew w

}

A Uo<

4.

4 1

I \\

e IV o

W

)

a MA \\

.~

m 9 n,

} v

-l-Y. \\.

4 -

2-

/J. /'

\\.

Jf vQ)",/.)

k. %.c _

, iiii g2

'4 O '

0-10e 10 - '

FREQUENCY-CPS FIGURE.lc

ACCELERATION RESPONSE SPECTRA B.

i i

i i i.. ii i

i

.iiii i

i 1 i i i R2UX

- 3D R/G ABWR REACTOR BLDG.

-- VP30150 - 3D R/B STHUCTURE-TO-STRUCTURE

- - VP3D150 - 2D R/B NODE 33 x

---

VP3D150 - 2D R/B+C/B + T/B RPV/MS N0ZZLE

22. DAMPING b

6.

G W

4 f

' Z'

' O H

b<m td h

'a 4-tij U

U 4

1 s

I\\

s

~ ' '

Q 4\\.)\\ i Si 2-e 9

v.-..

jg

\\.

r y \\

u~

k

/~

N

=

t i

1

't

'l f I f

I

'd I

i i t i i

B 3

i i i i I 8

2 10 '10e 10 10 FREQUENCY-CPS FIGURE 2a

.~

ACCELERATION RESPONSE SPECTRA 8.

i i

i i i ii i

...i i

R2UX

- 3D R/G ADWR REACTDR BLDG.

VP30150 - 3D R/B STRUCTURE-TD-STRUCTURE f

- - VP3D150 '- 2D R/8 NODE 89 x

--- VP3D150 - 2D R/B+C/B+ T/0 RCCV TOP 22 DAMPING J1 6.

rs tn Z-O s

.t-<

i tu l

4-ta

+

u U

4 J<m.

3

r. : - :

2. l

/ l,,, i Ld j.%

tt-

,f vs

./

\\ \\,

\\'\\..

i

% ' \\ :. -.

M' ' "3-R

• y. g :,,, 3... A.............

0.

3 2

10~1 10e '

10 -

l10 FREQUEtJCY-CPS FIGURE 2b

ACCELERATION' RESPONSE SPECTRA 10.

i i i i ii i

i i

i i ii.

. i > >>

R2UX

- 3D R/B ABWR REACTOR BLDG.

- VP30150 - 3D R/B STRUCTURE-TO-STRUCTURE

-- - VP3D150 - 2D R/8 NODE 95 X

--- VP3D150 - 2D R/B+C/B+ T/B R/O TOP 22 DAMPING 8.

en I

re tn z

O.

g F

6.

m to J

ia U

U j'

4 A

/

/1 4.

J m

\\ #: M n.

stJ11 S

[v v j,4 (U ' \\.

Q xn /, \\_

2.

e r...

./

(3 i

-- & A

%Nis,3

'\\ w-= v-eww=

0.

1 2

10'l 10e 10 -

10 FREQUENCY-CPS FIGURE 2c

ACCELERATION RESPONSE SPECTRA a.-

i i

i i...i i

i

> >>ii i

i i

i i i i i R20X

- 3D R/8 ABWR REACTOR BLOG.

~ - VPSD150 - 3D R/8 STRUCTURE-TO-STRUCTURE t40DE 33 X

- - VP50150 - 20 R/8 RPV/MS NDZZLE

--- VPSD150 - 2D R/B+C/B+ T/D 22 DAMPlf1G A

m i

6.

to tn Zo s

t-<

tr

.w J

4-Wo U

4

.r.

}\\

n Y.,

?

't\\

i! N,;!if

_fA\\.

\\

\\i N

1 a

2-y w.,.

q{\\

a

. Vn 3 /;-

-Qb-D.~em i,,i -

i i

i i:i e

i e

i i e i i 2

k 0-1 10 101 10 0

FREQUENCY-CPS FIGURE 3a l

_ _ _ _ _ _ _...... =

.. =,. - - - -..-

)lIlIl1I 2a 1

i E

i R

U T

.C i GU DR LT h~.__.

i BS R0 O1 T -

G

.CEX N

c~.

AR PI EU3OP RTBTM C

A RUEVD WRPC A

DTDC:

R ASNR2 t

T e

E b:

j C

S 1,,b.dm P

i t

.N.f

.i::.

E ilI

)

S S

l P

L.' !

i N

C O

's l

i P

l-I J

I Y

M E

E C

h S

A 3

N E

R A[.,

U R

Q U

E G

1 t

I O

R F

I F

T A

R E

L e

E 0

C

~.

8 1

/

C 1

A

+

B

/

C

+

BB9B

////

RRRR DD00 3322 000 555 111 XDDD 0SSS

. 2PPP RVVV 1

0

,1 e

6 A

Zowt<rw'wUU<

_<cFUW'n n

c Gn o

i t

t

i

)

ACCELERATION RESPONSE SPECTRA 10.

R20x

- 3D R/0 ABWR REACTOR BLDG.

4

- - - - WPSD150'- 3D R/B STRUCTURE-TO-STRUCTURE

- - WP50150 - 2D R/8 NCDE 95 X

--- VPSD150 - 2D R/B+C/B+ T/B R/B-TDP 22 DAMPING 8.

cn 4

r3 tn zo w

F 6.

tr I

l1 t:

__3 I,Y i til u

r-f3; 1

I, U

l i 1 [i.

A 4.

1 l8

.\\

1

-8

l. ;

bb.

N g 'l

,0,.

E

/du /

\\

2.

p-L

\\

k j@

%~tN

, k 0-2 8

1 2

10 -

10 10 FREQUENCY-CPS T

FIGURE 3c

l)Il

)(

ljll 1I l

i.

l!

l-l IliIi i

!1' 20 1

L E

~ _

U T

C GU OR L T L

,B S E

V L0 E

e O1 L

h R - x G

,T E

EN NR 1 DI i

, M.

OU5AP CT 1 R M C

GA 4-R UE D

A WRDB ET 0/2 R

ASnC2 2

T

>[

0 C

%. ; ~

1 E

\\

P 1+I.

\\

j

\\

S

\\

/./. '

E 4

S S

Q P

J gp.p C

t 0

\\

i.

f P

\\?v. '

\\,V i

Y I

i C

a S

p"

\\ N N

4 E

f

>k f

\\dl

\\,

U R

v E

E R

Q U

l_d

.\\

N E

G f

j! f:/ e I

G

\\,e.

R F

j F

I

A j I

j G A

y R

j E

-f L

"B E

C E

1

/

C T

A

+

B

/

C

+

DB8

///

CCR

+

, B

/DD0 C322 D-3 000 555 11 1 X0D0

,011 1 2BBB RUUU 1

0 m

0 5

5 01 0

2 5

7 0

3 2

icimn ZOsW<&wJ 4 J<eFowEm t

4i, - !

m

ACCELERATION RESPONSE SPECTRA 6_0 i

i e

i i i i.

4 i

i iiii i

i i

i i iii R20X - 3D C/B AB' R CONTROL BLOG J

- - UB10150 - 3D C/B STRUETURE-10-STRUCTURE

- - U81D150 - 2D C/B NSDE 181 x

--- UB10150 - 2D R/B+C/8+ T/D C/B TOP 21 DAMPING cn a

45 to tn Zo w

r<x j

w gj 3.0 ou<

J<

y, f

t-u At W

- / M s [A

's.

tt 1.s

.3

... v-y. ~,

..IY t.

g

. _ \\l k\\

j'

-j 0

10 ~ 2 10' 10' 1g2 FREQUENCY-CPS FIGURE 4b

1 l

~o w

_ a 1.I m

_ u ao II aw lal

9 J

-m m w

i e

>a w

].

o,us

( <I w

o

.- w wz h,y ze-o-aam<a "O~$5 i,

xaw a

v

/-

• 1 W D CD A

.g, ;

a-asa cr

<mzum p..-

,_.a

.c u

p%

-.1 w

5 ?-

h. *~~~**

?

g

-e

+s tn 5

(

m

/

v" Z

s

.s a

i a.

s.

-(*.

>u m

m x

s-m w

p.. ' fs w

a:

w F

, <A 5

d r

w a

m a:

c u.

m 1

W o

uh

~

u_

~

m s

u.

1 mmm sss uum m

sOOO umWN oi i i-

~

m owe m m. W-t xaco

-ammm m o. c. a CE > > >

I :.

II

?-

c.

CD D

O D

G **

e rv m

n o

m W

6_eS NOIlv83133Dv lV8133d5 F

t

ii ;

1ll!l, lilj

.1 I

i l

l

1. 0 1

i i

T T

E i

R U

T C

a GU DR LT

~

i BS 2._

LO OT Q%

Rx G i TE N

NR1 I

OU8PP CT1OM

'r' C TA RUE D A

WRoB BTo/1

\\

R ASnC2 c,}y.

\\

'0

,i T

1 C~

t 1

o E7i

\\

P j

m

~ i: ' -

\\

S i

\\

)

i :

E i

/

I l

c S

S P

i N

h C

t O

I J

.\\

o

~

P r

a Y

\\X C

b S

l E

r/ f.

N S

/

E R

v\\

U E

J

?,

Q U

R N

E G

O R

I F

I F

T A

R E

L 8

~

E 0

C B

1

/

C T

a A

4 8

i

/

C

+

8BB

/ / /

CCR DDD 322 D-

.. 3- 000 555 111 XDDD UJ33 2PPP n

RVVV 1

'0 0

5 0

5 01 6

4 3'

1 m n*

ZO~F<(EwJwou< s<rnd g e

t I

~

I 4

ii

.!2t i1 ii i

i1

i1!l,'lllil l[l,lf l

!1i!\\\\

Il jl!Iii!

I I

2 0

Y 1

E RU T

,C GU OR m,

L T L

,H S E

3 V

L 0 E

D1 L

F -

• G 4

,1 E

Ei 4

R 1 Dl rOU6AP 9

C T 1 R R C

G A RUE D

\\

A WRD8 i

ST D/2

\\

t\\

R A SP C2

\\

3 4

1 T

'0.\\\\.1 10 C

E S

'-vI-P p*/1 1

s E

S e \\ \\*\\.\\

S P

l C

t P

'm 0

Y S

C a

6 t

E R

E E

t!

R 0

UG N

E I

O q

R F

I F

T A

R E

f L

0 E

0 C

0 1

/

C T

A

+

B

/

C

+

0BB

///

CCE

, B

/D0D C322 D-3 000 555 11 1 XDDD 0SSS 2PPP RVVV 3

0 l

0 5

0 5

01 0

2 5

7 0

3 2

1 n

n < a4xsetcw so.

n*

zas.V cae i

t4i l

l

ACCEL.E RAT 10t1 RESPONSE SPECiRA 6.0 i

i i

i i i ii i

i i

a iiii e

i i

i i 1 a i R20X - 3D C/B ABWR CDt4 TROL BLDG

- -- VPSD150 - 3D C/B S T RUC T UPE -ID-S T RUC T URE

- - VPSD150 - 20 C/B NDDE 181 X

---

VP50150 - 2D R/8+ C/B + T/B C/B TOP 22 DAMPlHG cn i

4.5 m

tn zo e-+

F<

tr y

_3 3'0 j \\.

W u

3- \\

o y

4

/

tj g

?I.s

!\\

f k/

i\\

H O

.hlY

?.*

\\

\\

'h n.

1.5 uvw

/

-g/g 4

tn T

\\.

1

\\_-

\\

/>

.4

\\.

%g - %__ __.

~.

1-*--'-

I--#

~

~0 2

10-1 Ige 101 10 F RE QUEtJC Y-CPS FIGURE 6b l

_ _ = _ _ - _ _ _ _ _ _

(16)

Gli should perform SASSI analysis for tlie control building for tiie same site

~

conditions used for ilie Rit

Response

Control building will be analyred for the same number of SSI cases as the reactor -

i building h

..-l l

l i

\\

uo

-ll i

.l

.i d

t il

.I l

j

~l 3

....-.......... ~............................

t 2.3

. Design of Reactor fluilding, Reinforced Conciete Containment Vessel, and Control lluilding a

(1)

GE should provide the design loads and the design calculation of the access tunnels located in the wetwell for the pool swelling.

+

i

Response

These calculations will be made available in the February audit.

(2)

GE needs to revise the SSAR to change the definition ofimportance factor "1" in Section 33 I and table 3.3.1, and the wind velocity pressure distribution in table 3.3.2 x

Response.

Mark up's providM (3)

GE need to consider concrete cracking of the RCCV in the STARDYNE analysis of the reactor building to account for load redistribution. Also, considereation of sifTness reduction due to cracking of other parts of the reactor building should bc assessed.

Response

The STARDYNE model used for the static structural analysis of the ABWR Reactor Building (RB)is based on the uncracked properties.1lowever, the design -

seismic forces used for the STARDYNE analysis are the envelopes of the SASSI-analysis results which include both uncracked and cracked cases. :In order to study the effect of redistribution ofloads among the various structural components of the Ril due to the elTect of concrete cracking, the approuch used was:

(a)

Run a cracked STARDYNE analysis for the load case of global x-direction seismic using the same teduced stiffness values as the cracked case SASSI' analysis, The reduced stifTness values considered among the various -

structural components of the RB are as follows:

- RCCV Walls: 70% of the uncracked stiffness -

RPV Pedestal: 55% of the uncracked stiffness

- RB Outer Walls and Columns: 50% of the uneracked stifTness -

As the structure changes from uncracked to cracked condition due to seismic loads the seismic response changes The seismic load used for this.

cracked STARDYNE is therefore cracked case SASSI analysis results.

. 4

.a :

23 Design of Reactor Building, Reinforced Concrete Containment Vessel, and Control Building (1)

GE should provide the design loads and the design calculation of the access tunnels located in the wetwell for the pool swelling.

Response

These calculations will be made available in the february audit.

(2)

GE needs to revise the SSAR to change the deFnition ofimportance factor "1" in Section 3 3.1 and table 3.3.1, and the wind velocity pressure distribution in table -

3.3.2.

Response

Mark up's provided (3)

GE need to consider concrete cracking of the RCCV in the STARDYNE analysis -

of the reactor building to account for load redistribution. Also, considercation of sitTness reduction due to cracking of other parts of the reactor building should be assessed.

Response

The STARDYNE model used for the static structural analysis of the ABWR.

Reactor Building (RB)is based on the uncracked properties. However, the design seismic forces used for the STARDYNE analysis are the envelopes of the SASSI analysis results which include both uncracked and cracked cases. ' in order to study the effect c,f redistribution ofloads among the various structural components of the -

RB due to the effect of concrete cracking, the approuch used was:

. j (a)

Run a cracked STARDYNE analysis for the load case of global x direction seismic using the same reduced stiffness values as the cracked case SASSI -

- analysis. The reduced stifTness values considered among the various structural components of the RB are as follows:

RCCV Walls: 70% of the uncracked stifTness

- RPV Pedestal: 55% of the uncracked stiffness RB Outer Walls and Columns: 50% of the uncracked stiffness As the structure changes from uncracked to cracked condition due to seismic loads the seismic response changes. The seismic load used for this L

cracked STARDYNE is therefore cracked case SASSI analysis results.

r.-77p. re.

et-i-

-y-v y-f ety-K yy p

N

'7e P

s,-

gg 3-,,

men +-wy-n.-_w-mem

.<w-g p

g.

gmr*W-y,cg.-a.g-

-r--w.,-,

fr..-iyy*,

3rpye---p' 1-.y.9y---

w F-r

l r

I (b)

Compare the cracked STARDYNE analysis results with those of the design envelope seismic loads applied on the uncracked STARDYNE model.

.[

i The 3D. STARDYNE fmite element model of the RB structure is shown in figure

1. It is noted that only half of the structure is modeled because of symmetry and hence, half the total seismic loads are applied. The fmite element mesh at the base of the RB structure is shown in figure 2. The global x direction seismic shear

~

forces for the whole structure used for uncracked and cracked analysis cases are shown in figure 3.

1 The enveloped case and cracked case base shears obtained from the STARDYNE analysis as integration of element corner forces are compared below. It is noted -

that these values are for half of the whole structure that is modeled.

Base Shear (Lips) from STARDYNE j

i RB Wall RCCV Pedestal Int. Wall Total Emeloped Case 43932.27 40459.79 -6888 04 8656 42-99936.$2 Cracked Case 38506.85 34990.85 5475.30 7722.56 86695.23

[

Total Base Shear (kips) from SASSI t

Enveloped Case 102624.13 Cracked Case 88880.00 The above comparison shows that totalintegrated base shear obtained from STARDYNE equal the applied base shear force that was obtained from the SASSI analysis results. The comparison also shows that the results of the enveloped j

analysis case bound the results of the cracked case for the global base shear forces t

(4)

GE should consider the effects of out of-phase seismic responses (forces and moments)in the upper structures of the RB for.the static analysis and design.

Response

Attached report "A Study of Effects of Out of Phase seismic Shear Forces on the Design of Upper Structures of Reactor Building" provides Bechtel's evaluation -

results to show that the present analysis is adequate.

(5)

GE should update the SSAR to incorporate a description of soil pressure caculation for the embedded walls of the RB and CD.

.I

Response

This_ information was provided to the NRC.

r c}

-,,, -[ )

I,1.

m.l

,i

luu D

$3 Un G

CY

/

A /

//A

/ A/

A

• i l

N

/

llK

/

/

/A

\\

\\

n

/ 4

)\\

g

/

A/ _i

\\

\\\\

/

I

\\

\\

i i I

/

I I

i i

y

\\

\\

A

/

l.

I

/

/

/

/

g l

^l l

l I

I l

p o

^ '

0 E

E( l w e:ll

_n h

l

\\

.,r r

i g

7 g

k L i i i

1 1

Vf,

o 1,

_1

-i i

gg-

~. - -

I \\ \\ \\ \\

j

{ p n

\\

V\\

\\

\\

\\

\\

w

\\

\\

\\ h\\

_/ ' _/ ^.'

_/

a

~

n

\\

J\\

\\

\\\\ \\ hh

/

\\

NiG w y

\\

," y \\ \\

l

/\\

to

/, /

/A\\

/ / / / /

/

t h

/

/

/

/

\\\\

\\/ / / / /

/

2 4

/ / / / /

/

Q T

\\\\/////

/

/

/

/ A

\\V

/

V

/

/

/

/ /

/

/

/

/ // //

/ / / / /

/

V

/

/.

/ //

'/

/ / / /

/

/

/

/.

/ // / /

/ / /

/

/

/

g

/

/

/ //

/ /

/ / /

/

/

/

g cl P

\\5 (5

/

\\

n

4 1-g

- 1 q

3 1

h-N Q, %y(

i y %

%.. g m

...s,.

s s

i 4

(

Rccv k

' /

y 4%pngb k N

"""'2,4gggp'rjjhl

%g

{gy v

fl&o4E-7 ' N'"IT E -

E tem cN r MESH AT Tus ' 8ASE Or- 'R&~ 6T40cTORF y

,m y

+

ys+

1*. - -?-

J y-M e 'aw-w-

y e'

=

--7---------

lull RECv P1 DI*.I At

., o. _. _g _. ; _ __ g__ _ 3 _ g ____

f, l

i.itt AR Sill AR SilE AR 43 I

30 4g

_ l D G 31Cgr4 20 euw st. ops i_,

7 25 35 L

t 15 CRACKED 30 DEsica -'

20 bMVED I

E des tc,9 E

e p

j-p 25 g,,

j g

go

,-enveg3pg, o

g p' o

o i

_ 's

]

20

- CFAcKED w g

L-1

<r 4

q.

l e

10 l

us 15 8

un hs j

}

g s

3 s

g.

tis I

kJ l

kJ (Qggs 10 l

5 1

f l

0 5

I I

i O

1 1

I l

s

~

l

_3 _ 2Do t

- 25,000 e - 53900 s

-5 i

e t

= J00 t

35,oco _' 5p on o

=

3 1

~

-10 I

-10 I

-10 I-I 1

9.

1

'2 3

4

.5 6

0-1 2

3 4

0-t 2

3 4

5 6

00RCEt10**4 tons)

FORCEtt0**4 tons)

FORCEt10**3 tons) i Fmore-s-6 tor 3At.-

b Dm ecTion Sassic 3HeAc Forces.

- _ ~

L' (6)

- GE should describe, in the SSAR, the construction procedure for excavation,-

concrete pouring, etc. that will be followed for the constrution of CD close to RB.

Response.

The construction procedure for two closely spaced foundations is site dependent.

If the plant rests on hard rock, then special precautions are not neccessary. If the plant is located on a softer soil medium, both mats would have to be poured concurrently and in multillfi stages to prevent overstressing and cracking the basemats. The construction procedure for the basemat is site specific This item should be a COL action item.

(7)

GE needs to verify the adequacy of dynamic amplification factor of two for MSL break load on the MSL tunnel walls in the control building.

Response

The loads resulting from a MSL break can be idealized as impact forces with a rise time less than 0.01 secondsf The maximum dynamic load amplification factor (DI.F)is two for undamped single degree of freedom system regardless of -

oscillator frequencies. The use of DLF equal to 2 is conservative for the MSL tunnel walls.

(8)

GE should consider buckling of building slabs, such as CB slabs, due to the in.

plane loads from lateral soil pressure.

Response

Consideration ofinplane loads in the design of the CD slabs has been accounted for in the structural design. These loads are included in appendix 3H stress results.

(9)

GE needs to consider thermal gradient through MSL tunnel walls for the normal operation and abnormal condition.-

Response

The MSL tunnel thermal gradient or normal operations has been added to SSAR Appendix 311. The pipe break loads from a mainsteam line break or a feedwater-line break act over a very short duration. Therefore; the thermal gradient used for-abnormal operations is equal to the thermal gradient used for normal operations.

d

(10)

For the design of reinforced concrete foundation walls, GE should bend the tie rods by at least an additional 45 degrees (from 90 degrees to over 135 degrees).

Response

Design has been modified per stafTcomment. The changes have been incorporated into appendix 311 drawings (1I)

GE should provide studs to anchor roof to steel beams for upward force due to atmospheric pressure drop on the roof from tornado wind loads.

Response

Design has been modified per stafTcomment. The changes have been incorporated into appendix 311.

(12)

GE needs to check or design girder-to column and beam to girder connections for uplifl loads.

Response

Design has been modified. The changes have been incorporated into appendix 311.

(13)

GE should verify the adequacy of soil spring calculations in the static analysis and design.

Response

Attached report "A Study of ABWR Soil Spring Stiffness" provides Bechtcl? -

evaluation results to justify adequacy of soil springs used in STARDYNE model, (14)

GE should consider in-plane load in the design of building basemats-

Response

The in plane loads have been incorporated into the design of building basemats.-

- (15)

GE should include the presenice and inservice inspection requirements for liner and liner welds with reference to specific subsections of the ASME Code including

' description of the Code sections relating to the construction procedure in the.

SSAR.

"{

c

i il

Response

-A mark up of the ABWR SSAR was provided to the staffin the January 1993 j

ITAAC meetings (16)

Pipe break loads inside RCCV for feedwater and MSL breaks need to bc considered.

Response

Since mainsteam and feedwater line routings and support details are not fmal. Any i

design of RCCV features to protect the RCCV from pipe breaks of the mainsteam -

line or feedwater line would be premature. This item should be made part of the s

piping DAC, (17)

GE should discuss in the design calculations how the ACI 349 ductility requirements were implemented.

r

Response

i ACI 349 ductility requirements have been incorporated into the design.

3 (18)

GE should provide subcompartment pressure loads to Bechtel for the design of reactor building subcompartment walls.

Repense:

Bechtel was provided the subcompartment loads for their design and are currently fmishing up their evaluation of the subcompartment walls.

(19)

GE should provide design calculations for the removable block doors at reactor water clea up ;y:t= h-

• exchanger subcompartments.

-l itesponse:

-)

These calculations will be made available in the February audit.

9

)

-f

-. - ~

9w wvVmvwg iew--

-.p---twvm sin'-7 gr 1*tw'+'-Wy-V Y

'iT----Q v

g r-v,'sWewg y

- ' -g

w-P Tv t'iDvp ew-vv wi T rW igy vg'em ww-w v

'Tg-1

A STUDY OF EFFECTS OF OUT Ui FHASE SEISMIC SHEAR FORCES ON THE DESIGN OF UPPER STRUCTURES OF REACTOR BUILDING 1.

INTRODUCTION The structural analysis and design of General Electric (GE)

Advanced Boiling Water Reactor (ABWR) nuclear power plant structures were audited by the US Nuclear Regulatory Commission (NRC) during the week of October 15, 1992.

In that audit, the NRC pointed out that the effects of out of phase seismic shear forces in the upper structures findicated in Figure 1) should be considered for the design of the Reactor Building (RB).

The NRC is not certain about the adequacy of current seismic shear forces input to the three-dimensional (3-D) STARDYNE model for static analyses, which show all in phase from the top to the bottom of the structure (Figure 2),

resembling a profile that seems dominated only by the first mode. Although, these f orces, obtained from the 3-D SASSI soil-structure interaction (SSI) analysis (by the frequency domain method), do include the participation of all frequencies below 33 hertz, they represent only the absolute value of enveloped forces (at each level) from all soil cases.

Actual participation of higher modes in the upper structure may be truncated in this enveloping process because not only these shear forces, very likely, were not obtained from the same SSI analysis, but also they peaked at different times even in the same SSI analysis. Accordingly, these maximum shoar forces are conservative only for design of shear resistance of each individual story.'

However, they may not be able to produce critical global bending effects of higher modes in the upper levels of the RB.

The out of phase forces the NRC referred to are forces produced by higher modes.

Whether the seismic shear forces input to the 3-D STARDYNE model can produce bending deformation large enough to envelop the global bending effects in the upper level due to the contribution from higher modes (in any soil condition) is the main subject of this study.

2.

STUDY APPROACH As mentioned previously, the seismic shear forces input to the 3-D STARDYNE model was enveloped from all soil conditions (from 1000 to 20,000 ft/sec of shear wave velocity).

Among these soil conditions, it is reasonable to assume that the global bending deformation of higher modes in the upper structure can only be excited by high frequency components in the basemat motion..

These high frequency excitation become pronounced most likely in the hard rock condition. However, the ef fects of soil-structure interaction in -the brd rock condition are small in general.

Therefore, approximately, the basemat motion for this study can be considered the same as the control motion in the free field, and the rocking motion at the basemat can also be assumed negligible.

In this study, a dynamic time history analysis using the 3-0 STARDYNE symmetric model was analyzed for the East-West basemat I

motion.

The control motion in the SASSI analysis was input at the c

,_~

.--m-rm-

basemat.

In this dynamic time history analysis, all modes below 33 hertz were taken into account.

Accordingly, the higher mode ef fects on the global banding in the upper structure were included.

The maximum membrane forces and bending moments et the higher levels from this time history analysis are considered to be the critical among all soil conditions.

A comparison of the dynamic results with the static results would give a clear indication whether or not the static results are adequate for the design of the upper structures of the RB.

3.

COMPUTER PROGRAM r

In this study, the Personal Computer (PC) Version (May01/92) of STARDYNE computer program was employed for the analysis.

The analysis was performed on Compaq-486 33 mega hertz machine with 8 mega bites of random access memory.

4.

FINITE ELEMENT MODEL The 3-D STARDYNE symmetrical model (Reference 1) for the RB (symmetrical plane perpendicular to the Plant North direction) was employed for the dynamic time history analysis.

The soil springs of the model were removed, and the basemat was fixed in all directions.

The model is shown in Figure 3.

5.

DYNAMIC TIME HISTORY ANALYSIS The DYNRE1 option of STARDYNE program was used to perform the 3-D dynamic time history analysis by modo superposition method.-

In this analysis, all modes with frequencies up to 33 hertz were.

considered.

A uniform crP ical damping ratio of 0.07 was used for all modes.

In the mode superposition method of dynamic analysis, modal responses, of course, were combined algebraically.

The control motion used in the SASSI analysis was input to the fixed base STARDYNE model.

A total of 2000 time steps with an constant interval of 0.01 second were carried out, l

6.

COMPARISON OF ELEMENT YORCE8 The maximum membrane forces and bending moments of wall and slab in the upper levels were selected for comparison.

Elements-in this region and located at the east side of the RB were compared between the dynamic and static results.

Table 1 gives the comparisons showing that in all elements except Element 2059 the membrane i

forces on the wall from the dynamic analysis are less than the i

static results.

Element 2059 does not control the design of roof plan.

It should be noted that in Table 1, the vertical membrane j

forces, (Force-Y) are the direct response due to the global bending

at the upper levels, while the momenta about the x-axis (Homent-X) are local bending due to local effect only.

t There are three elements which have higher local dynamic moments (i.e. Elements 689, 842, and 2900).

When the reinforced concrete section of these elements were analyzed for the membrane force and roment, it was found that all the stresses in the concrete and reinforcements due to the dynamic forces are less than those due to the static forces, except for Element 2900.

The comparison of these stresses are shown in Table 2.

From tho static analysis results, the wall where Element 2900 la located is controlled by Element 2904, By comparing Element 2900 to Elethent 2904, it was found that the dynamic stress (2.492 kai) in the tensile reinforcement of Element 2900 is only 7 percent higher than the static stress (2.334 kai) in the controlling element.

7.

CONCLUSION From above comparisons,.the effects of higher mode participation (or the out of phase seismic shear forces) on the design of upper level structures of RB can be concluded as follows:

A dynamic time history analysis including all modes up to o

33 hz was performed in this study.

Out of phase seismic 1

forces associated with higher modes are therefore included.

Element forces obtained by the static method and the dynamic method are compared.

o For most of elements located in the upper levels, the membrane force obtained by the static method are higher than those by the dynamic method,-except one element on roof.

The dynamic analysis also gives lower local bending moments for most of elements except three, o

For those elements which have higher dynamic forces, analyses of stresses on reinforced concrete section snow that dynamic stresses are lower than their static counter parts, except in element 2900 which has a dynamic stress in tensile reinforcement about-7 percent higher than the static stress.- -This slight overstress problem can be overcome by adding additic.al reinforcements in the local area during the detail design phase.

From discussions above, the ef fect of the higher modes on o

the upper levels of the RB is considered adequately included in the static analysis.

1 i

REFERENCE 1.

ABWR, "3-D Hodel," Calculation No. C1-001, Rev. O, Job 18775-014,. July, 1992.

l

TABLE 1 Comparison of Element Forces Between Static and Dynamic Results Elam No Thickness static Analysis Dynamic Analysis (ft)

Force-Y Moment-X Force-Y Moment-X (k/ft) (k-ft/ft)

(k/ft)

(k-ft/ft)

El 41 - 49 689 3.28 178.8 3.0 72.7 51.7 **

692 3.28 199.0 9.4 102.9 3.5 696 3.28 129.4 3.7 75.5 5.0 El 74 - 88 798 3.28 66.8 32.8 32.1 7.4 801 3.28 63.4 12.7 43.5 9.1-805 3.28 39.9 3.0 38.7 5.2 I

El 88 - 109 835

1. 9 't 29.5 33.4 15.6 17.6 839 1.97 51.1 65.0 46.8 3.4 842 1.97 15.2 0.2 7.8 0.7 **

El 109 - 147 2900 1.39 8.9 1.1 4.0 9.2 **

2902 1.39 10.2 23.0 6.0 1.9 2904 1.39 20.8 17.2 19.0 1.0' Roof Plan 1975 0.98 32.1 20.6 34.8 1.6 2017 0.98 24.7 9.3 23.2 0.4 2059 0.98 1.2 7.4 8.1 0.3 **

Notes:

1. The local axis y is-in the vertical-direction, while the axis x is in the horizontal direction.

2. Dynamic force and moment do not occur-at the same time.

3. The snbol (**)

indicates that the= element has a dynaalc time history _ analysis result higher-than the static one, and for these elements, a stress-analysis of reinforced concrete section was given in Table 2.

L e

. - _.. _ ~ _ _ _. _,

TABLE 2 comparison of static and Dynamic stresses Elem static Stresses (ksi)

Dynamic stress (ksi)

No fc fs' fs fc fs' fs 689

-0.279

-4.474

~4.304

-0.211

-3.239

-0.309 842

-0.044

-0.708

-0.672

-0.026

-0.402

-0.277 2900

-0.046

-0.690:

-0.352

-0.182

-1.872 2.492 2904

-0.355

-4.322 2.334

-0.082

-1.267

-0.959 2059

-0.282

-2.081 3.856

-0.044

-0.667

~0.510 Notes:

1. An area of 6.0 sq in of reinforcement was considered in each face of Element 689,.While an area of 2.54 sq in each face for the rest of the elemt 'ts.

4

2. A compressive strength of'4000 psi was considered for the concrete.

+

3. The stresses fc, fs', and fs denote the stress in concrete, in compressive reinforcement, and in tensile reinforcement, respectively.

4. A negative stress indicates a compressive stress, while a positive, a tensile stress.

I P

4 q

. ~.

.,__.__,_..;_...,.__.1

_ __l...L f_ :

Il G.

Upper Levels Possibly Controlled by the Higher fr> R. C 2 t

Mode Participation 87.4o

- ?,s. n,,

- g s.ss g.7, g v

[

74 dl-74,9 7 i

5 6.74 I

Sb 73 s 3.16 9.n

8 "

39 2(

• 2.=ss 2 5 'W 7 $.C10

!!.96

-+

-_qg,,f3 S 27 J L

93 1 goys -;

a

- e, s 7 m.34 Ro.12 47 64

. - fl.12 r

---+

_ z..o R 7

_.,F, o g;,

- h. 6 *j i. <

-43 M

=

-52.33 i

upper Levels affected by the Higher Mode Participation Figure 1

1 i

R/11 R/II U/Is

..o

- 7 j

1 g-r

~1 ~ ~ t~- ~ i l-I Sil[ AR HIfJDIfiG I (IR', I (iri 49 4o 3 r, i

Ju I

w i

t g s,

.z

~ o

~

pg d

1 i

t.-

is

.i.

tu a,

t I

g j

i i

Y

,i 2

5

'l I*

-10 e-L1

'2

.3

~4 5

f>

U-S to 15 20 25 e-5.

Io l ',

Jo-i-

FORCEi10**4 tonst MUME t4 I t 10 * *S MOMENI(IO**4 i

=8 J

1

~

~ >

i i

i:

t i

t Figure 2-Seismic Forces Input' to the '3-D STARDYNE static Analysis

[

t t

. - _... ~.,_.

~ -. - - - - -

CALC k0.

Ci+001 tfv. C AffAChalki kt C Sullt t0 l

4 I

N D

b I

B B

B ms

~h r

cc c

N M

r Bc w

e v

2 i

G

~ [G!illi!ill!Ilh

• ilh

\\

~4 i

f

'r _

_ x

, x s

s l

l ill'I ll!Iillf HIllflilft_

y

\\

xx\\\\

\\ \\

• \\

g t x V//h \\ %\\\\ \\\\ \\ ' N \\

O'

, f)llii!IlffffIlh_

s i \\ \\\\N\\

x x, N N

si

'\\

N N x s NN\\\\ \\ s N sA N

""ffllllllllitill!HiHIHl/UHh N Vlx \\_ N

\\ \\-\\ 5 N

s i

\\V

/

"yN \\ \\f\\/ / ~

N/ \\ sj s

\\/

88 s

l l

l

8) lk

~ %

=

e

\\,/

N s \\

'[

\\

J l

Ys/

7'(

1 MMNy 1

'b

/

3f /

m ;Y 6

v r

/

/u,

/ < <

e l

'fg Y ;' $\\\\\\l - -.

E l

)

gj ji L

I 50 N'

i 9=..

'I i i i I N

g

/- i < / /

gg a

u f

y ',f.

,I

/ E

'l I ) t i I W

r ft T w== r g i\\

g \\

\\r:

L

(

I 1

\\

W.

)

1 l

94 s

[

_b Y\\

/\\ \\ \\ \\

~

l

{

/ /

/ '/. \\

\\\\ \\ \\\\

/

/

/

\\\\\\

\\ \\ \\ \\ -\\

M

\\ \\

'N \\/ \\ \\

\\ \\

\\

g, 9 s

?

l

[

t N,v%

/

/ /

oa l

s/ /

/

w l

\\

I

\\.

xf N

s x

x

\\ N/ N-4J

\\

\\,

C D[Ys]

i s

l*

i.-

1 l

M i

e

'f

'N i

L g

I

'h i

N t

b

=_--

J'._) O C l-l 7:

I

A STUDY OF ABWR SOIL SPRING STIFFNESS

.1.

INTRODUCTION The structural analyses and design of General Electric (GE)

Advanced Boiling Water Reactor (ADWR) nuclear power plant structurea were audited by the US Nuclear Regulatory Commission (NRC) on October 15, 1992.

In that audit, the NRC raised a question hbout the adequacy of modeling of soil resistance by the Winkler spring approach in the three-dimensional (3-D) STARDYNE models.

The purpose of this study is to conduct a atudy on the subject considering the coupling effects and to prepare a

satisfactory answer to the NRC's question.

To begin this study, a brief review of the ABWR modeling of soil spring was performed.

In preparing the three-dimensional STARDYNE models for Reactor Building (RB),

Control Building (CB),

and Radwaste Building (RWB), the overall soil stiffnesses in three translational directions and rotations about two horizontal directions were first calculated based on the Bechtel BC-TOP-4 D6 sign Guide (Reference 1).

In this design guide, a rigid foundation is assumed resting on a half-space elastic soil media The embedment effect is also considered in the calculations.

The translational overall soil stiffnesses were then assumed uniformly distributed over the entire areas of rigid foundation.

This soil stiffness over an unit area is generally defined as the subgrade soil reaction.

Based on the above assumptions, the soil pressure under the rigid foundation would be a linear distribution due to any loading.

The first approximation of this linear pressure distribution of the soil under the foundation mat is the Winkler's step function, in which the unit step function is applied to the tributary area to a nodal point in a finite element mesh.

In this approximation, each Winkler spring attached to a nodal point acts independently from the rest.

The NRC is concerned about the coupling effect of soil resistance.

Because a Winkler spring represents only stif fness of isolated soil columns under its tributary area of the foundation mat, the adequacy of neglecting the coupling ef fect between surrounding springs in predicting the deformation of foundation mat is the center of the question.

2.

STUDY APPROACH To answer this question, a triangular distribution shape function, instead of step function, between the surrounding nodal points was first assumed.

This assumption leads to a coupling stiffness matrix for a rectangular soil area.

This stiffness matrix is termed the consistent soil stiffness matrix, and the spring is called the consistent spring.

Two rectangular finite element models of foundation mat with the size as that of the RB basemat were created for this study, one of these two basemats is supported by Winkler springs, the other by

consistent springs.

Both models were analyzed for a loading condition approximately representing the actual dead load distribution.

The effects of the coupling terms on the RB basemat was determined by comparing the results from these two a' lyse;.

Two more analyses were carried out by reducing the thickness of the basemat from 18 feet (in the first two analyses) to 2 feet.

A larger couplit.g effect can be expected for this relatively flexible basemat.

Based on these comparisons, the adequacy of using Winkler i

springs in ABWR 3-D STARDYNE models can be determined.

3.

CONSISTENT SOIL SPRING STIFFNES8 A consistent soil spring stiffness matrix defines concentrated soil reactions at vertices or a triangular area or a rectangular area.

These concentrated soil reactions are statically equivalent to the total soil reaction integrated from the soil pressure over the surface area considered.

Consider a rectangular area with two sides a and b as shown in Figure 1.

Under this arca, the soil subgrade reaction is k.

If a unit displacement occurs at vertex 1, for simplicity and without af fecting accuracy, it may be assumed that a linear distribution of soil reaction, as shown in Figure 1, represents a reasonable soil behavior in response to the displacement.

In addition, a force of kab/12 is also assumed acting at vertex 1.

This corner reaction is required for the assuned (reaction) shape function to pass a rigid body test.

By considering the distributed soil reaction and'the corner force at vertex 1 as the loading, three non-zero reactions at vertices 1, i and 4 can be calculated statically.

It is found the reaction at vertices 2 and 4 are equal to kab/24, while-at vertex 1, kab/6.

Tb,-r *eactions represent coupling soil reactions at surrounding n(in 4.te to a unit displacement at one point.

Similarly, coupling N 1 reactions due to unit-displacements at other vertices can be found.

Finally, a-full soil stiffness matrix, K, can be setup as shown in Equation (1).

This matrix is termed the consistent soil stiffness matrix.

It is interesting to note that if +.he of f diagonal terms are lumped to the diagonal, the matrix becomes -diagonal, and it resembles the one for four independent Winkler springs.

4 1

0 1

l kab 1

4 1

0 l

K, = - - - -

0 1

4 1

l (1).

24 l

1 0

1 4

l 4.

COMPUTER PROGRAM

A brief survey of commercially available computer programs for foundation analysis reveals that Winkler springs are used most

widely, Although only a few commercial computer programs, such as STARDYNE and ANSYS, have the option for users to input a general stiffness matrix, none of these programs can create a coupled stiffness matrix for soil resistsnce such as that derived above.

In this study, the Per Nt. Computer (PC) Version (May01/92) of STARDYNE computer progras as employed for analyses.

This version was validated according to Bechtel Engineering Depe,rtment Procedure (EDP) 4.36 standards for engineering computer programs.

The analyses were performed on COMPAQ-480 33 mega hertz machine with 8 mega bites of random access memory.

5.

FINITE ELEMENT MODELS The RB reinforced concrete basemat, approximately 190 ft by 180 ft by 18 ft, as shown in Figure 2 is embedded 84 feet in soft soil with a shear wave velocity of 1000 feet /sec.

The Poisson's ratio of soil is 0.38, while its unit weight is 120 pcf.

With the assumption of rigid foundation mat on an elastic half space of soil, the overall vertical stiffness was calculated as 3,165,155 kip /ft (Reference 2) por PC-TOP-4 Design Guide.

By uniforr.ly distributing this total stiffness to the entire area, a subgrade

reaction of 92.5 ksf/ft is found.

k ca ue of symmetry, only half of the basemat was modeled as 9-sted in Figure 3.

The size of a plate element in about the i

same as the maximum element size in the ABWR 3-D STARDYNE RB model,.

symmetrical boundary conditions were specified along the edge from j

Nodes 1 to 22.

The basemat was also fixed in the X-direction at Nodes 1 and 241; every rotation about Z-axis was restrained because of no in-plane rotational stiffness was created for the plate elements.

A modulus of elasticity of 519,120 ksf and a poisson' ratio of 0.20 j

l were specified for the basemat concrete.

The reinforced concrete was considered as uncracked homogeneous isotropic elastic mater.ial.

l The normal unit weight of 0.150 kip per cubic feet was used, i

The identical geometry and material properties as described above were specified for the two 18-Ft thick basemat models, namely, the Winkler spring and the consistant spring models.

In the vertical l

direction (Z-axis), Winkler soil spring constants were specified in one model, while in the other model, a lower triangular consistent stiffness matrix was prepared for each area covered by a concrete plate element.

A Winkler spring constant was obtained simply by multiplying the tributary area of a node with the subgrade soil reaction.

A consistent soil stiffness matrix was created by using Equation (1) in Section 3.

Another two models were duplicated from the above two 18-Ft thick mod' ~ s with a modification of thicknesn from 18 feet to 2 feet.

In l

l

i summary, a total of four models were created as listed in the following:

Model 1 18-ft thick basemat with Winkler spring supports Model 2 18-f t thick basemat with consistent spring supports Model 3 2-ft thick basemat with Winkler spring supports Model 4 2-ft thick basemat with consistent spring supports 6.

LOADING A total RB dead load of 419,962 kips, which includes 92,592 kips for the basemat concrete, 96,390 kips acting on the exterior walls, 97,554 kips on the containment wall, 20,866 kips on the pedestal, and a remaining of 112,560 kips from miscellaneous items, was applied to the basemat models.

The basemat own weight and the miscellaneous dead load are uniformly distributed to the entire area, and the rest were applied as equal concentrated loads at the appropriate locations as shown in Figure 3.

It should be noted that only half of the above loads were applied to the symmetrical half models.

7.

COMPARISON OF ANALYSIS RESULTS Two analyses for the dead load were first carried out by using the two different basemat models.

The displacement and the bending moment along the edge from Nodes 1 to 22 (central section) are tabulated in Table 1 and clotted in Figures 4 and 5.

It is clear that the results from Winkler spring model are almost identical to those of consistent springs.

Two similar analysee were performed for a reduced thickness from 18 feet to 2 feet.

Table 2 and Figures 6 and 7 show the comparison.

Although the differences between the two models are larger than those for the 18-feet thick basemat, they are within 10 percent at both ends which are still tolerable in the design of the foundation mat.

The close comparison indicate that the coupling ef f ect between the isolated Winkler springs is small because of 1) the high bending rigidity of the plate element, or 2) the fine mesh size in the-model, or 3) the soft soil stiffness.

8.

CONCLUSION Based on the above comparison, it can be concluded that the coupling ef fect of Winkler springs in the ABWR 3-D STARDYNE models-are negligibly small.

Accordingly, their applications in these models for static analyses is adequate.

REFERENCE 1.

Bechtel Corporation,

" Seismic Analysis of Structure and Equipment for Nuclear Reactor Plants," Design Guide No C-2.44, Rev 0, August, 1980 (BC-TOP-4 Rev 4) 2.

ABWR, " Foundation Spring Constant," Calculation No C1-004, Rev.

O, Job 18775-014, July 29, 1992

D Table 1 comparison of winkler-springs and consistent springs for the id-Ft Thick 1 Foundation

=i Foundation Size 189.97 ft x 180.12--ft x 18.04 ft

' Modulus of Elasticity-3605 ksi (or 519120 ksf)

Poisson's Ratio 0.20 Total Vertical soil Stiffness 3,165,155 kip /ft Soil Subgrade Reaction 92.5 ksf/ft Dead Loads Basemat 92,592 kips-R/B Exterior wall 96,390 kips Containment wall 97,554 kips Pedestal & RPV 20,866 kips Miscellaneous 112,560 kips Total 419,962 kips Displacement (ft) at Central-Moment (kip-ft/ft) at Central' section Section Node No Winkler Consistent Elem No Winklar Consistent 1

-0.1174404

-0.1173525 1

762.7 767.3 2

-0.1188136

-0.1187390 2

1329.0 1337.0 3

-0.1205149

-0.1204529 3

1103.0 1110.0

' 0.1227580

-0.1227079-4' 801.8 807.7-4 5

-0.1256706

-0.1256319 5

378.3 383'.3 6_

-0.1292123

-0.1291044 6-

-146.5

-142.3 7

-O'.1331490

-0.1331311 7

-735.1-

-731.6~

8

-0.1370330

-0.1370237 8

-1036.0

-1034.0-9

-0.1394519

-0.1394493 SF

~1056.0.

-1054.0

.10

-0 1412223 --0.1412242

-10

-961.1'

-958.8 11

~-0.1418641

-0.1418684 11

-843.3

.-841.0.

-12

-0.1418667

-0.1418710-12-

-962.7

-960.3:

13

-0.1412301

-0.1412322 13

.-1059.0

-105720' 14

-0.1394657 '-0.1394632 14

--1042.0

-1039.0 15

-0.1370537

-0.1370446 15

-743.6

-740.2 16

-0.1331783

-0.1331607 16

-159.5

-155.4 17

-0.1292528

-0.1292254 17 1358.2-363.0' 18

-0.1257264

-0.1256882 18 769.5 775.1 19

-0~12283571 -0.1227863 19 1052.0 1058.0 20

-0.1206282- -0.1205672 20 1196.0

.1202.0 21

-0.1189986

-0.1189256 21-

'1141.0 1149.01 22-

-0.1177038~ -0.1176182 Note: _

The displacements are -in the vertical direction (z-axis),

while'the moments are about the x-axis.

Table 2 Comparison of winkler springs and Consistent spring for a 2-Ft: Thick Foundation Foundation Size 189.97 ft x 180.12 ft x 2.0 ft Modulus of Elasticity-3605 kai (or-519120 ksf)

Poisson's Ratio-0.20 Total. Vertical Soil Stiffness 3,165,155 kip /ft Soil Subgrade Reaction 92.5-ksf/ft Dead Loads Basemat 10,265 kips R/B Exterior wall 96,390 kips Containment wall 97,554 kips 1

Pedestal & RPV 20,866 kips Miscellaneous 112,560 kips Total 337,635 kips Displacement (ft) at Central Moment (kip-ft/ft) at Central' Section Section-Node No Winkler Consistent Elam No-Winkler Consistent 1

-0.2244124

-0.2370278 1

32.8 36.5-2-

-0.1013519

-0.1024860 2

57.1' 63.7 3

-0.0368905

-0.0327347 3

32.2 35.8 4

-0.0200685

-0.0154457 4-21.1 22.9 5

-0.0299362

-0.0266075 5

26.4.

28.1 6'

-0.0719396

-0.0711232 6

17.5 17.9 7

-0.1591330

-0.1619860-7

-94.1

-97~.2 8

-0.2412940

-0.2464139 8-

-100.8

-104.6 9

-0.2083540

-0.2113402 9

-24.1

-25.0 10

-0.1576429

-0.1563018 10 51.0 55.3 11

-0.1023780

-0.0974091 11-141.0~

148.7 12

-0.1023780

-0.0974091-12 51.0 55.3 13

-0.1576429

-0.1563019 13

--24.1

-25.0 14

-0.2083538

-0.2113403' 14

-100.8

-104.6 15

-0.2412941

-0.2464138 15

-94.1

-97.2 16

-0.1591332

-0.1619855 16

-17.5 17.9 17

-0.0719390- -0.0711223-17 26.4 28.1-18

-0.0299356

-0.0266064 18 21'. 2 '

22.9-19

-0.0200712

-0.0154484 19 32.2 35.7 20

-0.0369139

-0.0327629-20 56.8 63.31 21

-0.1015087

-0.1026763 21 35.3 39.4~

22

-0.2248164l -0.'2375298 Note:

The displacements are-in the vertical direction (z-axis),

while the moments-are about the x-axis.

!: !)u = 3 3 n -:

f

/

4 i

A.

/

y l

A N) '

/

I b id a~;mt

.4.1..T

• 1b-4' '

~

\\

/

4,m'4 a.o A//

n x m a,,,4 L.

+ p'j

\\1 l

T,1 I

'4 4

.> 4 4

6 4

4 4 'I l

!)7 9 i'

i

}

/i

1. '?

t&

gllg)t, it#

4

'505 l

.I

^

I Figura 1 A Retangular Area of soil Surface l

t l

.A.._a.4 s

2-1a.,__._$_.-

M n

.m+-5,.A

.-+EU,i.-u._-,1-*,M--a-m.E.-..*%

4g-.in

.J.pm..2 J-3 M. dP 4$

AtiA_

As.Md.

A A=

.a J4 J

M._w_a.

@)

-. 6)

- (Yh dh e-6h tu ch

[ll' _,

l...

_. d)l__-.._.f

_d

_. _. _il.. j,9 I,

Mr I

l I

l t

I lo a

"J(f t, 4-ri~d.1 4

s'

.s

,y n o r

n n

ze c.,r-i=ss n.= pr,g 2

-: m

. r.r-g ie -

p,

[.........

~ ~ '. gg

• I;j gg"1lf$2 hj p.,StdCi

$,.igj5U.

o ;._...

_j

'&k 4

/M

.~

i'NM#y1 gyy.

V

.g. -73 4 r&

gdyc$93p7,0.ttf1 i

b 4 3 M 'y hk

%NT ude t,

\\8,f88"I $j

~

E l NNA /

I i

s r_

4%pg1,6]

Tpggp

,en a

~

u gg 3 Of Mly ay

+

- g 4

p ls]

- t

- j g

~ g. o k

j

. 3 3

~

s

+

s -r j

t

. 1sflih

,TNihdi w

- (.,. y-g

\\

/./(4 v

f j

9 7 y

o

.: ] dp,f

.l

/

j rr i

t t

It.i

's a

' AF I:. V,>.N s

3' f

m).yTihr:

e d

gy.;$g!%g.j 1

.. g cc-uin

,,s k.

,_4(~.A Aj. fh%

i @ip p

.; 7 e, pp-1

, 0

_.q s

g U

p lJ.g FK-3 u]P4(n e wegi,-

ppij M t yJ flhen$

' t:

1g g4 n

m

=

2 j.

4c a-v.

ar

'*3E

• lY* E\\

- G-j ~~ ?s_

a 3

-T-. '.

~.

r

~-

0 w

• v'_

)

8 3k

_3' e

~

y k

00( f OU$f, 00t51' DiiDa 0000 is0WI-" l'$41~

"Mi

(@

00 K' r z

j e

/

w t

n-Z f.

.n.,.,w o lP

?ef ::~ t <.

-\\

21"

$)"ih a,< :1.w(&

y* to ll *: !? ly 15 If / 718 19 4 4 - '---

y 5 7 %

s r + v -+ + + + v v t v 11 7._. _ __

4 g'19 9 99 9-1 4 !

+=L.~1 h l!)ll?- @ W l:..

l 2

, ' ? l % : tb1 1h.? 9 10 l 11 ? N!

t f

,q :

i I

..t,

w

-*.4 1.; *,

t 1-

-1 i

?

g ej

. T3'

}

}

t

\\

\\{

l 131'.

d I

l af : 'l

, c'/ 2. -- $

I 7

i s! - j-a ga e;f c-/ :fy j

i

Jct t

-)

7,-

'j - g 1

-t i

o-4.

t

/ + i g-l la.

t i

,a1 p,

,I.--

s s( !/46, t.,e.,,

i 18 6 i

1 l

i i

~

jgj :i I

i i

i II:

/ -T..

ell /t/l 4

a[

~)iiufl/W.ilpl.1,1/n!i-}e/ tai /r :'/$/.'7'l"T/:1!/604/fl /62 /Q:/49:lsT'//fl/Q e

e e e

.e e e s e e

UI S feri w W all

)If9 A1 y A.

i I

Figure 3 Finite Element Model for the-Half of RB'Basemat i

P l

l-i l

e

ABWR STUDY 0: SOIL S 3RI \\ G STIFFNESS Displacement due to Dead Load 0.000

- 0.050

=

i 6 -0.100 g: 33 w

- 0.150

- 0.200 1

2 3

4 5 6 7 8 9 10 11 12-13 14 15 16 17 18 19 20 21 22 c

Nodes at Central Section

_,_ Winider spring 4 onsistent Spring C

Basemat Thickness 18.0 ft -

Figure 4 Comparison of Displacement at Central'section of the

18-Ft Thick RB Basemat t.

,r

ABWR S U JY 0: SOIL S39 NG S~lF:\\ ESS Bending Moment due to Dead Load 1500

.c 4

1000 i

X

~4j 500

=

t=

.5 0

=

1

-500

=

=

2 5 - 1000 2

- 1500 1

2 3

4 5

6 7

8 9 10 11 12 13 14 15 16 17 18 19 20 21 Elements at Central Section

+ Winkler spring

_q,_ Consistent Spring _

Basemat Thickness 18.0 ft Figure 5 Comparison-of Bending Moment at Central Section of the 18-Ft Thick RB Basemat

ABWR STUDY OF SO L SPRING STI:FNESS Displacement due to Dead Load 0.500 0.000

- g --

.- g. _

ET 5

$ -0.500

$.I C

- 1,000

- 1.500 1

2 3

4 5

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Nodes at Central Section -

4 onsistent Spring C

_._ Winkler spring Basemat Thickness 2.0 ft Figure s comparison of Displacement at central section of a 2-Ft Thick Basemat j

4 A3WR S U JY 0: SO L S R NG S~ F:4 ESS Bending Moment due to Dead Load 200

.c A

Ix 100 3

o 1

Id E - 100 E=

5

- 200 1

2 3

4 5

6 7

8' 9 10 11 12 13 14 15 -16' 17 18 19 20 _21-Elements at Central Section Winider spring

+ Consistent Spring Basemat T!.ickness 2.0 ft Figure 7 Comparison of Bending Moment at Central Section of a-1-Ft Thick Basemat

,w

_ _ _ _ - _ _