ML20056F923

From kanterella
Jump to navigation Jump to search
Summary of 930728 Meeting w/ABB-CE & Consultants from ABB-Impell,Duke Engineering Svcs,Inc & RP Kennedy in Rockville,Md Re System 80+ Sliding & Overturning During Seismic Events.List of Attendees Encl
ML20056F923
Person / Time
Site: 05200002
Issue date: 08/25/1993
From: Wambach T
Office of Nuclear Reactor Regulation
To:
Office of Nuclear Reactor Regulation
References
NUDOCS 9309010029
Download: ML20056F923 (71)


Text

/ U

["%g t UNITED STATES gu j

['.

, f NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 y' w , '

p

  • .* August 25, 1993 Docket No.52-002 APPLICANT: ABB-Combustion Engint.erang, Inc. (ABB-CE)

PROJECT: System 80+

SUBJECT:

SUMMARY

OF ABB-CE SYSTEM 80+ SLIDING AND OVERTURNING MEETING ON JULY 28, 1993 The U.S. Nuclear Regulatory Commission (NRC) staff met with ABB-CE and its consultants from ABB-Impell, Duke Engineering Services, Inc. (DESI), and Dr. Robert P. Kennedy to discuss the issue of steel containment sliding and overturning during seismic events and a number of open items related to the seismic modeling and anlysis of the CE System 80+ structures at the NRC headquarters offices in Rockville, Maryland, on July 28, 1993. -

The main purpose of the meeting was to discuss and resolve the issue of potential containment sliding / overturning and the need to provide a shear transfer mechanism between the internal structures and the steel containment, and between the steel containment and its concrete support. In addition, the

, meeting included a discussion of a number of issues raised in previous audit

meetings related to the seismic soil-structure interaction (SSI) analysis performed by ABB-Impell. A separate technical breakout session was held among the NRC technical staff, ABB-Impell, and their consultant, Dr. Robert P.

Kennedy to discuss the details of the containment sliding calculations performed by the applicant and the NRC staff.

Enclosure 1 includes a list of the attendees, and Enclosures 2A and 2B were used by ABB-CE team for its presentation at the meeting.

The meeting started at 8 a.m. on July 28, 1993, and concluded at 3:30 p.m. A summary of the agenda is as follows:

1. Evaluation of the potential for overturning and sliding of the System 80+

containment structure.

2. Technical breakout session among the NRC technical staff, ABB-Impell, and their consultant, Dr. Robert Kennedy, to discuss the details of the containment sliding calculations performed by the applicant and the NRC staff.
3. Resolution of open items from the audit of June 8-10, 1993.

ABB-Impell presented the results of their evaluation of the potential for overturning and sliding of the System 80+ containment structure. ABB-Impell considered four potential failure modes: overturning of the internal structure (IS) independent of the steel containment vessel (SCV), overturning of the IS together with the SCV, sliding between the IS and the SCV and I

9309010029 930925 8 PDR ADOCK 0520 2

{ u , g pgl

~

August 25, 1993 i sliding between the IS and the SCV combined with respect to the supporting concrete. The two stability requirements defined were that the line of action of the imposed resultant seismic force is located within the concrete support boundaries and that the coefficient of friction required for stable equilibrium is not more than 0.3 for the design basis earthquake and 0.4 for the review level earthquake. The stability analysis was performed in the time domain. The details of the modeling and analysis equations are presented in Enclosure 2A. The earthquake levels applied were a review level earthquake of l 0.6g, used for the determination of seismic margin, applied at the ground surface, and the design basis earthquake of 0.39, as specified in the CESSAR.

The acceleration time histories for the " worst case" soil conditions were determined by comparing the acceleration profiles of the IS for the three control motions (CMS 1, CMS 2, and CMS 3) for the various soil conditions specified in the CESSAR. Based on the analysis performed, ABB-Impell  :

determined safety f actors of 1.0 and 1.2 for the review level and design basis earthquakes, respectively.

The staff commented that ABB-Impell had computed the seismic moment causing the potential sliding based on the orientation of the worst case resultant horizontal earthquake force rather than the worst case moment. In addition, the NRC staff had performed independent approximate calculations based on envelope acceleration profiles which indicated significantly lower safety margirs. Subsequent to various technical discussions, the staff recommended that the standard design should include provisions for the shear transfer mechanism between the IS and the SCV as well as between the SCV and the supporting concrete, and that the combined license (COL) applicant should be given an option to delete the shear transfer keys if it can be justified for a specific site. It was also decided that a separate technical breakout session should be held among the NRC technical staff, ABB-Impell, and their .

consultant, Dr. Robert Kennedy, to discuss the details of the containment  !

sliding calculations performed by the applicant and the NRC staff.

a A technical breakout session was held among the NRC technic, staff, ABB-Impell, and their consultant to discuss the details of the containment sliding '

calculations performed by the applicant and the NRC staff. In this session, the participants reviewed the impact of using maximum acceleration profiles and the worst case vertical earthquake effects rather than the stability analysis in the time domain. The technical breakout session resulted in the following conclusions:

1. ABB-Impell will repeat the sliding / overturning analysis by evaluating the stability in the plane of the maximum seismic moment about the center of the SCV instead of the plane of the resultant horizontal seismic force. ,
2. Instead of the time phasing of the vertical earthquake, ABB- Lnpell will use 40 percent of the maximum vertical earthquake and apply it in the upward direction in all cases. ,

i

August 25, 1993

3. Positive shear transfer mechanism such as shear keys will be provided to resist the earthquake sliding force taking into account the friction force based on a coefficient of friction of 0.3.

ABB-Impell presented the results of their studies in response to the open items from the audit of June 8-10, 1993. The following four items were discussed:

1. ABB-Impell performed a study to evaluate the impact of the variation of the poisson's ratio from the value of 0.4 assumed in the CESSAR on the vertical seismic response. ABB-Impell found that the vertical response spectra were significantly impacted when the SSI analysis was performed using poisson's ratios of 0.3 and 0.47 and damping of one-third of that used in the horizontal SSI analysis as committed in the CESSAR. However, ABB-Impell contended that the current thinking of the industry is that the P wave damping used in vertical SSI analysis should be equal to the S wave damping used in the horizontal SSI analysis. ABB-Impell performed a study to evaluate the impact of using poisson's ratios of 0.3 and 0.47 using the same damping in the P wave analysis as in the S wave analysis.

As shown in Enclosure 2B, the impact of the variation of poisson's ratio in this case was insignificant. In addition, ABB-Impell contended that regardless of the damping and poisson's ratio used in the certification application, the site can be considered as qualified as long as the vertical motion at the surface as developed by the COL applicant for site-specific conditions is enveloped by the System 80+ surface spectra included in the CESSAR. Based on the results of this study, the staff concludes that the COL applicant must address the impact of the poisson's ratio and its variability on the vertical seismic response.

2. During the June 8-10 audit, the staff had raised the issue that the cut-off frequencies used in the SHAKE computations impact the strain ,

computations. The staff had requested that the applicant should evaluate I the impact of frequency cut-off on the high frequency computation in the soil column for the soft sites where a low cut-off frequency was used.

ABB-Impell stated that a cut-off fregmy of 40 Hz was used in all SHAKE analyses in order to satisfy the criterta that the cut-off frequency chosen is high enough within the frequency range of interest and consistent with the frequency content of input motions.

3. ABB-Impell performed a study to address the staff concern that deep soil sites of relatively stiff soils may not be covered in the range of columns investigated by the applicant. The study evaluated two additional soil column cases. In the first case (D-1), the soil column was extended to a depth of 400 ft. In the second case, an evaluation was conduced of a soil column in which the shear wave velocity varied from 1000 fps to 3000 fps over a depth of 800 ft from the ground surface. The details of the shear wave velocity variations are shown in Enclosure 2B.

The results of the study indicated that the surface motions from both of these soil column cases were within the envelope of the surface spectra used as the design basis for System 80+.

_4_ August 25, 1993

4. During the June 8-10 audit, the staff had indicated that ABB-Impell needs to show that Winkler soil spring and the uniform soil spring derived from a finite element analysis are compatible for a governing SSI case. ABB-Impell re-analyzed the Winkler sensitivity model with applied loads from case B-1 for the CMS 1 motion of the SSI analysis. Based on the results shown in Enclosure 28, ABB-Impell concluded that the Winkler representation is adequate for the purposes of static analysis of the superstructure and the basemat.

ABB-CE personnel committed that ABB-Impell will reevaluate the sliding based on the orientation of the worst resultant seismic moment. They also stated that positive shear transfer mechanism will be provided to resist the earthquake sliding force taking into account the friction force based on a coefficient of friction of 0.3. The standard System 80+ design will include a positive shear transfer mechanism, but the COL applicant will be given the '

option to eliminate the shear key mechanism if it can be justified for a specific site.

t ABB-CE personnel will present the results of this meeting to their management and provide any feedback to the . staff before the August 16, 1993, meeting ,

between ABB-CE and the NRC management. i (Original signed by)

Thomas V. Wambach,-Project Manager Standardization Project Directorate Associate Directorate for Advanced Reactors and License Renewal Office of Nuclear Reactor Regulation

Enclosures:

As stated cc w/ enclosures: i See next page DISTRIBUTION w/ enclosures:

Docket File PDST R/F DCrutchfield  !

PDR PShea TWambach DISTRIBUTION w/o enclosures:

TMurley/FMiraglia WTravers RBorchardt BDLiaw, 7D26 GBagchi, 7H15 SMagruder TBoyce JNWilson  ;

ACRS (11) TEssig RPichumani,7H15 SAli, 7H15  ;

MFranovich OFC LA:PDST:ADAR PM;PBST:ADAR- SC:PDST:ADAR NAME PShe C d TE'mdickfz TEssig N DATE 08h3' 08/25[93 08/f/93 OfflCIAL RECORD COPY: MSUM0728.IVW f

ABB-Combustion Engineering, Inc. Docket No.52-002 cc: Mr. C. B. Brinkman, Acting Director Nuclear Systems Licensing ABB-Combustion Engineering, Inc.

1000 Prospect Hill Road Windsor, Connecticut 06095-0500 Mr. C. B. Brinkman, Manager Washington Nuclear Operations .

ABB-Combustion Engineering, Inc.

12300 Twinbrook Parkway, Suite 330 Rockville, Maryland 20852 Mr. Stan Ritterbusch Nuclear Systems Licensing ABB-Combustion Engineering, Inc.

1000 Prospect Hill Road Post Office Box 500 Windsor, Connecticut 06095-0500 Mr. Sterling Franks U.S. Department of Energy NE-42 Washington, D.C. 20585 Mr. Steve Goldberg Budget Examiner 725 17th Street, N.W.

Washington, D.C. 20503 Mr. Raymond Ng 1776 Eye Street, N.W.

Suite 300 Washington, D.C. 20006 Joseph R. Egan, Esquire ,

Shaw, Pittman, Potts & Trowbridge 2300 N Street, N.W.

Washington, D.C. 20037-1128 Mr. Regis A. Matzie, Vice President l Nuclear Systems Development ABB-Combustion Engineering, Inc.

1000 Prospect Hill Road Post Office Box 500 Windsor, Connecticut 06095-0500 i

i

4 I

LIST OF ATTENDEES JULY 28. 1993 lLAMI ORGANIZATION B. D. Liaw NRC G. Bagchi NRC R. Pichumani NRC S. Ali NRC T. Wambach NRC D. Peck ABB-CE

5. Ritterbusch ABB-CE L. Gerdes ABB-CE R. Kennedy Struct Mech Consultant l S. Esfendiari ABB-IMPELL N. Kathrotia ABB-IMPELL T. Oswald DESI J. Johnson DESI I

Enclosure 1 j i

1 I

m.  :

. P1 OF 2$ l EveLosueE %A I

SYSTEM 80+ ALWR  !

CONTAINMENT OVERTURNING AND SLIDING EVALUATION l t

l i

l 1

I 1

NRC MEETING 1 July 28,1993 I

1 ABB Combustion Engineering System 80+

l

EWL }A : P.S .

Meeting Agenda July 28,1993

1. Overview
2. Modeling

~

3. Analysis Equations
4. Earthquake Levels
5. Analysis Results ABB Combustion Engineering System 80+

. E tatL 1 A . P. 3 OVERVIEW OBJ8ECTIVE EVALUATE THE POTENTIAL FOR OVERTURNING AND SLIDING OF THE SYSTEM 80+ CONTAINMENT STRUCTURE 67 ft R-100 ft SCV IS (Crane Wall)

El' Elev 91.75 ft

\\ 7/ _

ei v s f N Local bearing area  ;

l l

Sketch of the IS and SCV ,  ;

1 l

l ABB Combustion Engineering System 80+

1 EMCL Q A : P' 4 .

OVERVIEW (Continued) i FAILURE MODES  :

The potential failure mechanisms include: j (1) overturning of the IS Independent of the SCV, l (2) overturning of the IS together with the SCV, (3) silding between the IS and the SCV and  :

(4) sliding between the IS and the SCV combined with the supporting concrete. j l

I i

l i

i 4

5 ABB Combustion Engineering System 80+ f

y  ;

, F.N f L $ h : 05 L

MODELING STABILITY REQUIREMENTS  :

- The following model represents a sphere with the center of gravity (eg) located at the center, and forces FH and FV acting at the cg.

- A reaction force, normal to the sphere and collinesr ,

with the imposed resultant force, will develop.

- Stability is maintained as long as the line of action of the imposed force is located within the support boundaries.

scv c , sev c- ,,,,,, ,_ .

~~ s

\ a .iani ro,=

A I i a "- i s i . . .. c,,,,,,,

o la de su boundaries. 5 i

~

ABB Combustion Engineering System 80+ .

t

i enn- y :p.c .l 1

'i MODELING (Continued)

, r

- The next model considers a sphere with a moment t applied at the cg, as well as forces FH and FV. This l' moment would account for eccentric forces due to differences in accelerations imposed on each mass.  ;

SCV  ;

i M  !

... .. >,. FH l l '.l t FV l 'j., I l ' .,  ;

Resultan I - - ', ., (

Force '

, ' -. , i s

.,. j........ s

v. - * '

PT.B *

. s

~ ~ ~ ~ ~ . . . . . . . . .g-l PT.A  !

Location of reaction forces shifts from PT. A to PT. B  !

iH order to maintain stability.  !

i

. The reaction forces now include a frictional shear i force in addition to a normal reaction force. . l

.l i

i ABB Combustion Engineering System 80+ l

- -[

.-- ._ , - _ _ _ p

weL f A : p. 7 MODELING (continued)

The normal reaction force is no longer collinear with the imposed load. The location of the reaction forces will shift such that the imposed moment can i be resisted.

The frictional force is a function of the normal force and the coefficient of friction.

Stability is maintained as long as the frictional force can be developed without sliding or slippage and the location of the reaction force is within the support boundaries.

l l

l l -

ABB Combustion Engineering System 80+

2

evet % p.s MODELING (Continued)

- The SCV and IS are analyzed using a two-dimensional model.

- A stick model was used to represent the mass distribution for the SCV and IS. See Figure 2-4.

- The forces acting on the Individual masses are transformed to the center of the SCV. The resultant of the horizontal forces and the vertical axis defines the " failure plane" in which overturning and sliding will tend to occur. See Figure 2-2.

- The analysis model is shown in Figure 2-3.

ABB Combustion Engineering System 80+

P EA>CL Q A : 9 1 MODELING (Continued)

Y FH = I Fxncos e + I Fynsin 9 Fa1 j r Fa2 Fy2 Plas rias ladividual forc es Resultam (Ceekood)

(*d shsht occumuen a from SCV Csow) (ignanas acescuat.as trem SCV Canw)

Fa1 k

i r-t> y D' 1 J P Fb1 F2JL = , =

g. 2 fFh2 Resuhant Force Resuhans Force Elevados Elevotion Eleve tien In div uluel F orces Resultast Fares Fm coup;e et Cenw of SCV

( gno its acce.ucsues frata SCV Cenw)

Figure 2-2 )

1 1

Resolution of Forces  ;

1 ABB Combustion Engineering System 80+

i EHtL QA : P. J o i

MODELING

)

(continued) n +Z j R-100 ft  !

SCV  !

r r

t M

i

' FH

(

l ' ., l +X FV l, 'l

.., t Resultan I--- . I Force j,/' .,.,

, 4 ,-------. ,

'.'. ' )s ', FN

.,~~.. ..p -

g 3

Moment causing sliding - M Resultant force is due to inertial and deadweight effects.

Normal reaction force - FN Tangential reaction force - FT - (coefficient of friction)FN ,

cx defines the location of the reaction forces. [

defines the orientation of the resultant force.

  • Figure 2-3 l

Analysis Model .

l i

ABB Combustion Engineering System 80+  :

EWL QA : f 11 MODELING (continued) l Elev. 257 460 9 ,

l Elev. 250.97 449 l l ) 529 Elev. 208.122' l

l I ) 526 Elev. 189.429' Elev. 221.28' 437 h '

l l ($ 521 Elev. 165.7 l

! I I 518 Elev. 143.586' Elev. 174.37' 425

() 515 Elev. 126.73*

( ) 512 Elev. 114.133' Elev. 122.8' 413 $ ,

! ( ) 509 Elev. 104.51' Elev. 91.75' 401 h..........

g g 506 Elev. 88.089'

() 503 Elev. 70.193' Internal Structure Figure 2-4 Stick Model of the Internal Structure and SCV ABB Combustion Engineering System 80+

~

EtJPL KA : 9-12 ANALYSIS EQUATIONS

- The seismic accelerations for the nodal masses are obtained from the time-histories used in the generation of the System 80+ response spectra.

- Seismic inertial forces for the three global directions are determined for each nodal mass. .

- The direction of the resultant of the horizontal inertial forces is determined. This direction defines the " failure plane" in which overturning and sliding will tend to occur.

. Determine the horizontal inertial forces in the failure plane and transform these loads to the center of the SCV.

- Normal and frictional shear forces are generated at the interfaces between the SCV and the adjacent concrete.

- The location of the resultant force, as defined by the angle a, necessary to maintain stability is determined by the static equations of equilibrium.

ABB Combustion Engineering System 80+

(

ENCLQA. P. 13 3' ANALYSIS EQUATIONS (Continued)

The development of the analysis equations follows:

Nomeno:sture:

mn mass at node n g gravitational constant Ain Acceleration (expressed in terms of a fraction of g) at node n in the direction I (includes any scaling factors to account for differences between the ground surface ZPA and the rock outcrop ZPA)

Fin seismic inertial force at node n in the direction i Fhn horizontal force at node n acting in the failure plane FH total horizontal force acting in the failure plane Wn weight at node n  :

W total weight FV total vertical force including inertial and deadweight effects e angle between the individual component forces and the failure plane cose FX/FH sine FY/FH Hn distance from the center of the SCV to the nodal mass I M moment about the center of the SCV i FT tangential reaction force at the SCV surface  ;

FN normal reaction force at the SCV surface i a angle defining the location of the reaction forces l D angle defining the orientation of the resultant force due to FH and FV

$ a-p coefficient of friction (maximum of 0.4 for seismic i margins and 0.3 for design basis) l R radius of the SCV l l

ABB Combustion Engineering System 80+

i suet y: p. m .

ANALYSIS EQUATIONS  ;

(Continued)  ;

It Fin = (m n)(g)( Aln)  ;

i FX = IFxn l

FY = IFyn l l

FZ = IFzn ,

FH = [(FX)2 + (FY)2]0 5 i W = IWn i F V = FZ + W Taking moments about the center of the SCV in the failure '

plane, M is obtained.

M = ((Fxn)(cose)(Hn) + (Fyn)(sine)(Hn)}

Using the static equations of equilibrium, the following relationships are obtained: l (FT2 + FN2)D 5 = (FH2 + FV2)o.s l FT = M / R  !

i

- I i

-i l

, x ABB Combustion Engineering System 80+

r.

EM t'L "L A . P.15 ANALYSIS EQUATIONS (continued)

Substituting for FT,

[(M/R)2 + FN2)]o.s = (FH2 + FV2)o.s (M/R)2 + FN2 = FH2 + py2 FN2 = F H 2 + FV2 - (M/R)2 FN = [FH2 + py2 . (M/R)2]o.s

= FT/FN

= (M/R)/[FH2 + FV2 - (M/R)2]o.s The angle, a, is calculated using p and 4:

p = tan-1[FH/FV)

By taking the summation of moments about the location of the l reaction forces, M-(FH2 + FV2)o.sRsinc = 0 sin $ = M/[R(FH2 + FV2)o.s]  !

$ = sin-1{M/[R(FH2 + FV2)o.s]}

Using the relationship between 4, a and p, a is calculated:

$=a-a= +$

ABB Combustion Engineering System 80+

EntL A A : P- I(, -

ANALYSIS EQUATIONS (continued)

The maximum at is defined by the elevation of 62.75 ft. The SCV center is at elevation 157 ft.

Rcosamax = 157 - 82.75 amax = cos-1[(157 - 82.75)/R]

I.

amax = co s-1(74.25/100) amax = 42 degrees By engineering judgement, limit aallowable to 80% of amax in order to ensure that the reaction forces are distributed over a larger area under the base of the interior Structure.

a allowable = 0.8a max a siiowadie = 33.6 degrees ABB Combustion Engineering System 80+

ENCL 3A: P IJ EARTHQUAKE LEVELS Seismic Margin Earthquake

- The 0.6g CMS 3 earthquake is used to evaluate overturning and sliding.

The 0.6g CMS 3 earthquake is taken as two times the 0.3g CMS 3 rock outcrop spectra developed for the design basis. The rock condition as well as the other governing soll cases are evaluated.

Soll Cases In the generation of the design basis seismic spectra, a 0.3g motion was applied at the rock outcrop.

The resulting ground surface motion obtained from the soll column studies (which are used as input into the SSI models) is often higher than the input at the rock outcrop, i

For the seismic margins analysis, the overturning and sliding capacity, expressed in terms of peak '

ground. acceleration, is required. This acceleration is taken to mean the ground surface acceleration.

I ABB Combustion Engineering System 80+

eneu ;L A : P it ,

l EARTHQUAKE LEVELS ,

(Continued) .

. Therefore, in order to obtain the nodal  !

accelerations for the various soll cases (associated ,

with an equivalent 0.6g peak ground acceleration) for the seismic margin earthquake, the soll zero period acceleration (ZPAs) are normalized (scaled) to a 0.6g level at the ground surface.

- In performing the overturning and sliding checks, the normalizing factors based on the horizontal ZPAs are used to scale both the horizontal and vertical accelerations.

- Table 1 shows the scaling factors for various soll cases.

- Cases CMS 3 Rock, B1 and C1 were selected as the governing cases for analysis.

Design Basis Earthquake  ;

. The acceleration time-histories for the " worst -

case" soli conditions are considered. The " worst case" soll conditions are determined by comparing i

the acceleration proflies of the IS for all three control motions.

i

- By judging the acceleration profiles, Cases CMS 1 ,

Rock and C1; and CMS 2 Rock and B1.5 were analyzed for design basis.

I 1

ABB Combustion Engineering System 80+

r sw eL. } A r P.IG EARTHOUAKE LEVELS (Continued)

Table 1 Scaling Factors for Normalization Soll Case Ground Required Scaling l Surface Margin ZPA Factor

B1 0.420 0.6 1.43 B1.5 0.460 0.6 1.30 i B2 0.500 0.6 1.20 l

l B3.5 0.520 0.6 1.15 l

B4 0.390 0.6 1.54 C1 0.380 0.6 1.58 C1.5 0.420 0.6 1.43 i C2 0.260 0.6 2.31 C3 0.390 0.6 1.54 A1 b.420 0.6 1.43 Rock 0.300 0.6 2.00 Scale Factor = 0.6g/ Surface ZPA  !

l ABB Combustion Engineering System 80+

t t

E A4* L M: 9 20 j i

e EARTHOUAKE LEVELS .! i (Continued) o i  !

i  ?

TABLE 5-1  ;

SURFACE ZPA NORMAUZATION OFINTERIOR STRUCTURE ACCELERATIONS j j

(CMS 3) r W7ERNAL STRUCTUfE B1 31.5 B2 33.5 84 C1 C1.5 C2 C3 f830K A1 ,

HORZ 8LMA22PA 0.420 0.460 0.500 0.520 0.390 0.380 0.420 0.250 0.300 0.300 0.420 HORZ F ACTOR (Nots 1) 0.714 0.652 0.600 0.577 0.769 0.789 0.714 1.154 0.769 1.000 0.714 VERT S1 MACE 2PA 0.240 0.300 0.390 0.540 0.490 0.260 0.370 0.480 0 500 0.200 0.220 I

t VERT FACTOR (Note 2) 0.833 0.667 0.513 0.370 0.408 0.769 0.541 0.417 D.400 1.000 0.909 Notes: 1. HORZ FACTOR = 0.39HORZ SURFACE ZPA

2. VERT FACTOR = 0.2pVERT SURF ACE ZPA f

NORMAUZED SOILTO ROCK ZPA ACCELERATION RATIOS '

(Factor bmes soil spectral accesabon desped try rock spectral acconeraton) t

DlR ELEV B1 31.5 St B 3.5 54 C1 C1.5 C2 C3 fCCK At j X 50 3.14SD1.067 0.612 0.431 0.579 0.816 0 652 0.696 0.413 1.000 0.729 l X 70 L147 t t.154 ' 0 644 0.4 54 0.686 0.949 0.702 0.756 0.490 1.000 0.893 X 92 01.0052 0914 0.552 0.414 0.653 0.864 0.586 0.648 0.473 1.000 0.867 X 105 0.920 0.787 0.500 0.392 0 810 U.80'3 0.533 0.560 0.440 1.000 0.828 .

f X 114 0.860 0.763 0.458 0.377 0.560 0.734 0.491 0.505 0.405 1.000 0.776 t

X 127 0.832 0.739 0.445 0.357 0.538 0.705 0.479 0.477 0.374 1.000 0.738 X 144 0.777 0.696 0.407 0 411 0.532 0 618 0.475 0.402 0.340 1.000 0.661 f X 166 0.740 0.660 0.381 0.444 0.580 0.536 0 441 0.332 0.292 1.000 0.607 f X 189 0.707 0.610 0.319 0.4 38 0.577 0.511 0.392 0.252 0.224 1.000 0.508 X 210 0.716 0604 0.200 0.422 0.570 0.508 0.374 0.219 0.198 1.000 0.490 f Y 50 0.838 0.911 0.702 0.654 0.641 0.992 0.745 0.731 0.477 1.000 0.721 [

Y 70 0.759 0.796 0.612 0.593 0.593 0.860 0 649 0.638 0.439 1.000 0 698 .

< Y 92 0.786 0.654 0.468 0.477 0.492 0.668 0.539 0.497 0.375 1.000 0.630 [

! Y 105 0.796 0.563 0 427 0.422 0444 0.603 0.488 0.419 0.331 1.000 0.612 F

I Y 114 0.797 0.546 0.397 0.401 0.454 0.591 0.475 0.386 0.309 1.000 0.605 Y 127 0.796 0.557 0.431 0.399 0.408 0.583 0.470 0.371 0.290 1.000 0.621  ;.

Y 144 0 692 0.583 0.359 0 420 0.419 0.543 0.477 0,310 0.255 1.000 0.635 l Y 166 0612 0.545 0.424 0.397 0.416 0.556 0.450 0.253 0.225 1.000 0.603 ,

l Y 180 0 661 0.458 0.362 0.358 0.542 0.489 0.418 0.202 0.194 1.000 0.542  ;

0

! Y 210 0.686 0.452 0.344 0.352 0.541 0.490 0.409 0.184 0.180 1.000 0.520

2 50 0 858 0 683 0.495 0.504 0.461 1.012 0.768 0 338 0.322 1.000 1.018 Z 70 0.739 0.572 0 418 0.425 0.374 0.772 0.658 0.296 0.253 1.000 0.936 I

Z 92 0.649 0.503 0.378 0.376 0.318 0.745 0 589 0.265 0.253 1.000 0.867 j

Z 105 0.586 0.492 0.357 0.046 0.286 0.744 0.546 0.250 0.237 1.000 0.833 I

2 114 0.567 C.501 0.348 0.335 0.275 0.748 0.527 0.242 0.227 1.000 0.811 I

l Z 127 0.546 0.502 0.340 0.325 0.266 0.746 0.516 0.234 0.229 1.000 0.795 Z 144 0.513 0.513 0.327 0.316 0.242 0.757 0.490 0.217 0 208 1.000 0.733 f 0.493 0.514 0.309 0299 0,238 0.777 0.469 0.203 0.197 1.000 0,697 F

l Z 166 l Z 189 0.486 0.532 0.300 0.293 0 264 0.796 0 458 0.194 0.186 1.000 0.671 l Z 210 0 475 0.550 0.304 0.293 0.264 0.805 0.459 0.192 0.188 1.000 0.661 . [

l i RATIO e 107 Y Y N N N Y N N N N Y Medan X 0.805 0.718 0.426 0 413 0 573 0.662 0 477 0.439 0.357 1.000 0.700 [

l Medan Y 0.739 0 552 0.410 0 400 0.449 0 569 0 473 0.341 0272 1.000 0 608 ,

i l Wadian Z 0.529 0.508 0.334 0.320 0.265 0.752 0.503 0.226 0.217 1.000 0.764 t

l i

! ABB Combustion Engineenng System 80+

> i

. , . - , -,-v r,,,, , . .~ -.. -- - - -, . . , , - . . . . . - . - -- -

i EUPL .M : P. L j EARTHOUAKE LEVELS (continued)

! wooeosw-esoe-meenww-nemewoooms e ese t -neonneeommew eeenen-esonneomem e wee j q u c h a s s e c e w s s e h n s e m e m s h e c e m e m,m m N eNN oooooooooooooooooooooooooooooo o ooo 1

esen seu,

  • O.

8

. . o, ocoo------ooo-------oooooooooo g,se8 8 oe-e

, NeOQ ,C , , , , wkheeqq%ee n

soe8menons w uunwen s

g o

,,N

--o

,a meeeums-wennawesees.ne-emooowe s n u s. w m nws unnow-es-co-enewww.essnomeno e u oooooooooooooooooooooooooooo n " * * *

  • q q N N * * * * * * " " " " ' * ? ? *. *. ? " o"9 o ':o " 'o o

T o

o e e m e, uswwoom

- e m c e e - s, nocwonvesowssssume eceoNwewn-nNo3we N mee n n,as e men N nan N  ?""

U ooooooooooooooooo T * *. * *. S. ". ". " " *. *. ooS. " " " " o" " " o o" oo " " o". oo ". " o o o o ooo

@ oseom e see O

m.nweenem-N-moneoneenseoNe-onemw unwwsemnon-esenmeo--seen

-wwwwwwnnnnmemeewwnennnneeeneme a

w noe wwn uddo66666666666666o666666ao6666 6 666 CA wun o 4

1X weNNSoom*ommeemsenecoo r.usew-nneeonsowesmm-oo 8a C**N*

enne ~

om8 me

    • O N< U*********'NNS*******00 dooooooooooooo66oooo44444466oo o o64 4w2 m eab .ee---.n.e~e.~

wwensene-mewwNomsemmoenomeennwN

.ua--own. -e-e- -

n

.nn sne W * ***

J H N o * * * * * * * * * * *. * *. * * * *

  • M e n % % * * * * * *. "

g O oooooooooooooooooooooooooooooo o ooo a o-CO pHm g .eumwoommuncuoovemo-oweow-omoe-e-wenwooeemeneenes--se-nnnesee n

u eme nnw w emeemmwwenwmenneemeneee e e,m ne C ddddddododododddddedddddddoddd d ddd 9s CU w wenneemnuomeosmeomsweemse--mem

, oee W~ eeeesoaccese-wmeenu,cemeeseums ~

v.

mon w=m owrwwre.n-nneem.=w9 " w r w. e n v = = = = w. n n oooooooooooooooooooooooooooooo o ooo m.senewowneenmoneows-sweeonssnw oeN-o

- e m e m e .e .n .N .N .n .e e. o v e s n N e e Neseemwwwseeeemesem oeesnwnN m a w N

e ocu eeN mee Coddooddoodododdodoodddoddooddd d doo meseNNeoMoeNwe-NeN NmnemNwmome e W -wm sp NesmN e en oo-w NN Nm s -s o c e m e e e e . .e e. me.vommesemmeenemennnnnnem-s .w e een ,

-oooooooooooooooooooooooooooooo o ooo j F $5[ ,

> e snewo e seeao avsweea <

oon,o-mwee- ss3

.3 , ------uoou,o-avem ggs doon o-uwee- 3 3 '3


u ------~ ,

x>N n.

gxxxxxxxxxx>>>>>>,>>>~~~~~~~~~~ j .

I ABB Combustion Engineering System 80+

1

ENCL Q A : P 22.

EARTHQUAKE LEVELS (continued)

-ome.e eeomese nngemom soseemome o non 20* 282EEEE0EKEEcE0'zsquSQ ;803 ; @ R.ooo ococoooocooocococooooooooooooo o emme,o-memeesNnNeNesNo-,m* * ,,e SON $$5NEeS$e$NESO$seNENe$$goesopSE e No$

co66oodooo66cocoo6666odo664oo6 o 66o eseenonssmo--nnomewomounnennae n ee.

O$$$Nh$$$EbNSENNSNNNN$ENE$NE$5h h $ " ,~ N oooooooooooooooooooooooooooooo o ooo o o v

e n e w m e n c' s N eo-oo-mn-oeo-seose-n n ecN omene mmmomeowooe woesesesome e e n. ,e U " N ". 9 "

  • C O
  • N ".".".". 9
  • 9 * * * ? V ' ' ' V n M ".*.*

/ 0 "o.".

000000000000000000000000000o "o. ooo

c. onN,-e e n e m e N e s ,- e n c. ennuses,em.-es

- e N e n m e e m s o N N. e o on, O - nem ,~ o ~n - ~ ~

. . .~ . n- .nnnnnnn=messeesee ~

-e~

,ns F- Uo66oo6c66eo6ogo66o6oodo66odood 6 ago 4 ^

oea W N anu g ,,,,un_nang,ume,,n,,,uomu,, n A AX 60ESESEESEeOOOS$$EEEEEE0EENS E eEO N4 ooooodo6oooooooo6oooo6ooodo444 o co6 DW2 g cQ -ossnoe.,oue,eu- noueu.somous, , oes O'3h 5GEEEEEEGE02EE%8%nSEE%EEEES$0$2 SH oooooococooooooooooooooooooooo o

% UEE ooo mCO

%M $M Hg g

emeevnewemn--menese-swemmemo-e m"?$8800.07%SEEEEEEEEEEEEffE*R$

n $

a Nwe

$$E q {O mCm Goooooooooooooooooooooooooooooo o ooo O 492

, F CO u-em--.cne.se. ...n....-.ee.ne . --.

W'. useomenem-meswoomsmemmsmooowose nn

?

sse

    • n Tnnn TVnn o h moooooooooooooo

'

  • m. * *
  • n ' ? ' ? ' ' ' o* o o *o *o o o o o ooo ***oooo o ooo I eennnomesem-mess-em* ens d m oogogsom-eve-neemsomosse=ex-moonnsame o =,

s nse mes

-cooommseN esesseemmmmenesssssss G e4'46ooooooooodo6ocoo6o6ododoo o oo6 4

oooeo- e-..nmeremes sen,e.neuem m num

[%

X o mmmmmeeeeemmemeesr,ns m-@ w eMnnN ns -w oN d em e m o,,m e ,te ,n m e, e, m, m, n

e nN W me.

-c o c o o o o o o o o o o o o o o o o o o o o o o o o o o o o ooo

  • 23 EEs i m .s,emo e .o -E w %w >

d Cryo,s...oawec

-- ----~

CC" o-mwee-


n Ryo,s. -wwee-


u 3

1 5 - T g- s 4.c ,

x>N N =

j a

gxxxxxxxxxx>>>>>>>>>>~~~~~~~~~~ .

j k

'st:>

d }".4 Q 64 ABB Combustion Engineering System 80+

eueu qn : p ts t

. EARTHOUAKE LEVELS (continued)

AoNenoonNooomenw-enooveN ome=,se N-nnesnoo-ne nee-nneweNnnnwwesen ememewooNnwwweewoonnwwwwwwween 6o666oo6sdoo66o60 sod 666666odo6 e me.

- oesmemoeoen8wh osoww mooeon een-comensweosse oeweoenwus-weeunea e sno Sh e

< * * * *

  • N *. * *
  • N *
  • N N N * * * * *. *. S. * * *.
  • 9 *. ?

oooooooooooooooooooooooooooooo o ooo emes-emmewomem-NoneemsneonomEom sneumsu-e---ooms eeneseeonce meesssssseseoemesseeweessssowe ne ,-

e-w ums see dooo66o66o66o66o66oo66666ooooo 6 666 ow ee-eo-oowsneneoo-noesemomeos e --s n- o--s-sn-wewemucu-omoo--No,es e e,o u n n ? * ? " n N N N ' ? *.n NNNNnn '.n n N N " nnn oooooooooooooooo n " o o o o o ooooooooo

'.* ? o ooo ennenenencenNee-onsensowen--on n -me uwesew-cosemmonam-seneo-unnowes e neo

  • ""N NN9999 9 9Nn u n n o. n n n N N 9 9 5cooooooooooooooooooooo"?"NN oooooooo o ooo O

O ..unw mommse-mn*sws-ew-nowoooneseemo w n-ww nnww e-esnoooneo---oono e

w noe use

- -mecemew w w e.s w e m m e w w w e e s u s s w e e n e new U 66cd66co6666666odo6666od6666od o 666 OC1 W N s su, e num 3 4 sococco,Sno,ons-..e-unu.

ow nwname-u eemeeno-nueue-neow- w nee z1X

-N4 ue*NNN S******************NNNN**

ooooodo6ooo6oooo66oo6o6ooo6ooo 6 ooo

  • WE z

OC3L

<s e-,seoomuns -sone,oenu-wwone-o , uou oH emees owonooone-mennoon-oommecom - emo N m??*He********************???'.9 O oooooooooooooooooooooooooooooo o ooo

't 3 d MCO Wgem .Nes,es,ome-nnenweeosmon-nece-e nw w w w e s e e c c e e n m e c o c c e e e emeeeeem museos-eme-wee-ownum= e e e e e w ,e a

e uss oow a n e mee mCm m oo66666oo6oo6cocoo6odooo6co666 6 o6o

<Os CU concenow ocewo enesnecem useonne s one w~ nos -oosee-onn o emeeneyw ww eesepeymewnmuomooensome,ensw wwwww, ynn a

w som mew oooooooooooooooooooooooooooooo o ooo l

1

, wens-s-ns,oo-eme,sweeneen-e-oce ecuswe suncoo-sn oceommemoeawwoecce,em o

n e-e swo

-ooommessosooossseecew w s ese Go66o6ooo66446oo6oooo6o66666odo o o6o mennwoe,oeo-oew-coe-ee-so-ne-e asusoce-e-wsooosessnemoooos-en o

e nee eme I

l m * * * *.*

  • N N N * **O***NSN#5 '9????'9n N *** i oooooooooooo-ooooooooooooooooo o ooo '

> oom ewswooo oon cesweso oomp oewsweoo i EEE -

jeg p o -Newe


~ ggp o

-Nwee---~ gg Nwee-


~ 2

) vsk 0 4 eee X>H XXXXXXXXXX>>>>>>>>>>NNNNNNNNNN ABB Combustion Engineering System 80+

G^W L. ^}.A : P. g y l EARTHQUAKE LEVELS ->

(Continued) j Additional Margins in Spectra

- An estimate of the margin that exists for the CMS 3 rock case is obtained by comparing the spectral accelerations at the fundamental frequencies of the IS and SCV as follows:

interlot Structure Direction Freauency Actual SooLG1ER Tarcet Soectra Ratio
  • H1 (X) 9.54 hz 0.69g 0.63g 1.1 H2 (Y) 8.56 hz 0.73g 0.63g 1.15 VERT (Z) 23.25 bz 0.32g 0.32g 1.0 Steel Containment Vessel Direction Freauency Actual Soectra Taraet Soectra Ratio
  • i H1 (X) 5.2 hz 0.67g 0.63g 1.06 H2 (Y) 5.2 bz 0.65g 0.6 3 g 1.03 f-VERT (Z) 11.9 hz 0.5 g 0.42g 1.19 This ratio is equal to actual value divided by the  ;

target ~value and is a measure of the adoitional margin. .

- At this point no credit for this additional margin is ,

taken for the overturning and sliding checks. ,

ABB Combustion Engineering System 80+

.n -i.i---..is -mm-----_-- m-.i--m,mem._m-.,q,mem-y _.-

o

$ System 80+

g CMS 3 Rock Outcrop Spectra

& Horizontal Motion 1, 5% Damping

1. 6 0

-~~ ~ -

4

~-

O

...s.~.

I --~'

m

-4 4 _. . -

tu 3 .. . I .. f .

p _ _. _ _. ._ ..

_ .. ,I ..

. _.. l 8 'g l '

@ 1.20 -

^ .. ..

o d.

\

- o> .

l

c. _ _. . _ . .

.O__

m 16 0.00 - -- - - -- - - - - - ~ ~ " - '- -

y g .

A/%A - .-.--- - - . .

Nlh j g; Y

y 0.00 -

~

/ -- -

- ~ ~

k *--- - - - ~

9 0.40 - - -~ ~ ~ -~/ _ ~~ '

~~ - - ~~ ~ ~ '

N o < p f h .,f74 I o z 2 o 0.20 ..

.o p

g O y C A g

. ..r _.

o M o- # I r

w, u

a: 0.00 _f o

E 0.10 1.0 2 >

C 10 10 -

t Frequency (Hz) 3 Figure 3.7 b

. Synthetic Time IIistory III spectrum vs. Target spectrum for CMS 3 (q o ,

b8 0 hm it s ') ta b

J' a<

o SYstem 80+

CMS 3 Rock Outcrop Spectra

$g Horizontal Motion 2, 5% Damping 1.60 '~ ~~ ~ -

1.40

-- - - - ~ '- -' - - - - - ~ ~~- - '

g o E 1.20 - - -- - - - - - - - -

l7 -- - - -

& 3 99 . . . - - .. .

.. l .

~

c . - . . . ... - .-. - - _ - . . . - - .. . ..

O 0.80 - - - - - - ^ - - ' - - - - --

i@

G y . . . - .

%%g . ..

x= e- e s 4-- "'-

0 60

- ~ ~ - -

o g =

w =. o t 4 . .

0.40 - - -- - - - ~ ~ ' - ~~- - - - ' - '

s

.. . _ _ . t. . . .A Y

I -# ~'"

m

- - --- ~~ ~--

0.20 b

6 o 62 -. - . . .- . . .. . - .. . E

!  ! 0.00 '

a "

2 9

" g 0.10 1.0 10 10 b

[ Frequency (Hz) t Figure 3.8 -

Synthetic Time IIistory II2 spectrum vs. Target spectrum for CMS 3 N A

o ,

$)A b 9 -4 t,r oD 2

., , . , - - . , v . . - . -- - -

.- a a a System 80+

- - - -'- -

1.00

.n L

- - - - - ~

M C ... , .

l.. . . .

O ii 0.80 (D L . . . . . .

fD

-y TwC u @ - - - -~ - '- -- '

0.60

- - ~ ~ - -

o Iyr- ,,,,,, o o

o g..

a 4 . _ ..

e g

. .~. .

2 0.40

- - - - ~ ~

U( A A.- .

s

' /

lo 5 0.20

,g . . _ _. _.

n,

/p<

t m 8 0.0 0 t-en o

' ' 2 9 8 c 0.10 1.0 10 10 c 3 E

Frequency (Hz) ,

w Figure 3.9 Synthetic Time IIistory V spectrum vs. Target spectrum for CMS 3 9

4 i

WJ }

l DJ .4 m D9 SN I

l

sa e t. %A : p. 2s ANALYSIS RESULTS ,

d The and a for the various earthquakes and soll conditions are:

and a for Overturning and Sliding Condition Earthauake/Soll g a (dearees)

IS Alone 0.6g RLE/ CMS 3 Rock 0.395 30.8 IS and SCV 0.6g RLE/ CMS 3 Rock 0.388 ajL2 IS Alone 0.6g RLE/ CMS 3 B1 0.358 20.5 ,

IS and SCV 0.6g RLE/ CMS 3 B1 0.313 21.7 IS Alon~e 0.6g RLE/ CMS 3 C1 0.321 19.3 IS and SCV 0.6g RLE/ CMS 3 C1 0.308 19.6  :

IS Alone 0.3g DB/ CMS 1 Rock 0.253 16.,.S IS and SCV 0.3g DB/ CMS 1 Rock 0.252 18.4 IS Alone 0.3g DB/ CMS 1 C1 0.165 12.0 IS and SCV 0.3g DB/ CMS 1 C1 0.160 11.0 IS Alone 0.3g DB/ CMS 2 Rock 0.221 15.7 IS and SCV 0.3g DB/ CMS 2 Rock 0.224 15.8 IS Alone 0.3g DB/ CMS 2 B1.5 0.239 13.3 IS and SCV 0.3g DB/ CMS 2 B1.5 0.227 14.7

- The critical values of and a are underlined. Based on a review of these values, it is evident that the governing earthquakes are 0.6g CMS 3 rock for seismic ~ margins and CMS 1 rock for design basis.

ABB Combustion Engineering System 80+

F r.

E Ne L .38 : P. 2 9 l

ANALYSIS RESULTS (continued)

SUMMARY

OF RESULTS By scaling the earthquake level, it is found that the following safety margins correspond to limits of or a bcIng attained:

Earthouake/Soll Maroin Limit 0.6g RLE/ CMS 3 Rock 1.0 sliding 0.3g DB/ CMS 1 Rock 1.215 sliding The earthquake level at which the safety margin is equal to 1.0 is given below:

Earthouake/ Soil Level (o) Maroin Limit RLE/ CMS 3 Rock 0.6 1.0 sliding DB/ CMS 1 Rock 0.36 1.0 sliding ABB Combustion Engineering System 80+

1

/ -!

/

t P. I of 37 .

3

. Eucto s u g e A8 ,

1 l

l l SEISMIC ANALYSIS OF SYSTEM 80+ NUCLEAR ISLAND AND NUCLEAR ANNEX STRUCTURES l RESOLUTION OF AUDIT OPEN ITEMS -

l I

I l

l Presented By:

ABB Combustion Engineering l July 28,1993 ,

ABB COMBUSTION ENGINEERING SYSTEM 80+

ENf L }6: P. 2.

SENSITIVITY OF VERTICAL MOTION TO POISSON'S -

RATIO OF 0.4 June 8-10 Audit Item: (12)

QUESTION:

In the SSI analysis, sensitivity of the surface motions to three effects needs to be addressed:

(a) Effect of depth on degradation in soil stiffness and dampingfor deep soilsites (b) Degradation sensitivity to soil types (i.e., sand or silt)

(c) Effect ofpoisson's ratio (0.4 assumed) on verticalSSI analysis RESOLUTION:

  • Items 12(a) and 12(b) were agreed as COL checks as long as the surface motion falls under the envelope of system 80 + motion
  • Soil Column studies were performed for case B-3.5 with Poisson's ratio of 0.3 and 0.47 respectively
  • Results attached ABB COMBUSTION ENGINEERING SYSTEM 80+

4 esec 2s : p. s < r N

l w _

o A

ewQ Q

&n => -

=* ji cn ei os o D c

$ s.

nW

=s o wA ao - e

~

sg *o $

it -

E o g m's e s.is m Me o.

e

.= =

g- o .s 'o .9 %

= ,

o -

@ b=

n!& et oc m

.i..... ... u

\ '

n o

O $eC  :

=

'. 6 88

~z .e,

........... ........- gg 8 pw o o o o o oo R N v D m o-a E m ~ed. '

. . . . . . D

% Q$

b e o o s C V M V a C 3

-k -

0 0 6 D 'oC ii il o @

11 o o W o k$ 1 m .*o

.ii .% 7-

  • w' o i

l 4

m E% ss s*

e

$ .% c C

%c l

c c c y e e o o s e n o -@ .c 'O d .E .E .E

  • o @y eee E -o i

t e

20 s

i e o o = 0 v 5w e ~,

9 e ,

9 o o o o o O o

W v to to  %

-l

2) - acepng punojo mojag lydaG

9 Peak Vertical Acceleration - g 0.2 0.3 0.4 0.5 0. 6

0. 0 0.1 ' ' ' '
  • 0 ,

, .~

o a

1 9

k 20 l q

l

s

@ j .. '

m 40 -

g -

I e .

e o

S sl gg _~ ,

h ,

S!

e o

m -

o Case B-3.5 S ,

~

00

~

Poisson's Ratio = o.40 n q t

  • - - - Poisson 's Ra tio = 0.30 .

p Poisson's Ratio = o.47 y e.

100  ?

Fig.' 2-2 Calculated Peak Vertical Accelerations for Case B-3.5 Using Various Poisson's Ratios and Considering Damping Values for Propagation of P-Waves to be 1/3 Darnping Values for Propagation of S-Waves Juriv

~ **: P. s~

Frequency - Hz  :

O. I 1 10 100  ;

2. 5 .

l Case B-3.5 Poisson's Ratio = 0.40 i h = = = - Poisson's Ratio = o.30 m

as Poisson's Ratio = 0.47 -

2. 0 l' B i

- I i l l  ;

i -

e ~

o ,

~

9 l ',  ;

i . j _

e ..

5q3

1. 5 . 1 l:

,  :. ,k t, <, .

o . ,

o '

a T -

s

' !q 3

l . i '

b o

ll

, j '

h '.

8 C) i d q 1. 0 ,

e p, .g i

  • k-Co - '

s ,o' hk l 's -

l t -

r a

l'i[j-- j ~

0. 5 ,'i sI  !

I a Nw -  :

,s

, 3

~

^

Damping = 0.05 -

~

d O. 0 ,

Fig. 2-3 Comparison of Vertical Spectra Calculated at the. Ground Surface for Case B-3.5 Using Poisson's Ratios = 0.40, 0.30 & O.47  ;

and Considering Damping Values for Propagation of P-Waves to be  ;

1/3 Damping Values for Propagation of S-Waves

- n,

GN f L 'SS : P . (,  ;

Frequency - Hz O.1 1 10 100

2. 5 ... . . . ...... .

- All Cases Considered in CESSAR Case B-3.5 using Poisson's Ratio = 0.30

'j 'g, 2.0 q -

i t .

~

9 .

2 /

5o 1. 5 N, ll, 0 -

I i c l '., f f l(

~

i L a

% j 4 G, i

1. 0 fp

%q -

? .

) -

j

~ -

.g h [

f f j

A/[

k[ ,s4{NI T.-

I

[

(jir J

Damping = 0.05

~

~

0. 0

^ -

Fig. 2-4 Comparison of Vertical Spectrum Calculated at the  ;

Ground Surface for Case B-3.5 Using a Poisson's Ratio = 0.30 '

with Spectra Obtained for all Cases in CESSAR 2llk",'Z':ll'YL'd l,'lo7 .'l'E" lllll" ~inR"lllll" " "' Ju!y I

~ " ' "

~

Frequency - Hz l 0.1 1 TO 100

2. 5 3

.o . . . . ...., ,

  • All Cases Considered in CESSAR ,

Case B-3.5 using Poisson's Ratic = 0.47 2.0 '

\ -

' ~

J !

t: ~ +

~

l

@ / i '

.S? 1. 5 f, I l l 8

1 1 o /

b I

K -

I l'  !

ll I

~

j Q I -

l E -

0 \

O d a ti ,( i\ --

q 1. 0 h f, U .

g -

g l[\

s ' -

Iv

,h t  ; -

{l

~

V f

U 'E~

n'm 0 o~

?yy~ i

~

Damping = 0.05 -

f'd ~

0. 0 Fig. 2-5 Comparison of Vertical Spectrum Calculated at the Ground Surface for Case B-3.5 Using a Poisson's Ratio = 0.47 ,

with Spectra Obtained for all Cases in CESSAR Note: damping values for propagation of P-Waves were considered = ff3 the damping values for propagation of S-Waves for eII the cases shown in this figure.

July .

m Frequency - Hz  :

0.1 1 10 100 2.5  :

- Case B-3.5

~

Poisson's Ratio = o.40  ;

--- Poissons Ratio = o.30 Poisson's Ratio = o.47 ,

l ,

2.0 -

en .

C .

.O -

~ .

e .

5e 1. 5 .

O .

O .

T .

~ .

t . .

e 1.0 a .4

,g to

. i ,

.y ..

aO v ,, -

.c s  : .

/;A .  :

L' *? ' ^

O. 5 . pp i-q ,1

- e ',' -

') i J

/ \ Qlg*f - -

lll -

Damping = 0.05 -

l

d

0. 0 .

4 Fig. 2-6 Comparison of Vertical Spectra Calculated at the Fo'undation  :

Level for Case B-3.5 Using Poisson's Ratios = 0.40, 0.30 & O.47 -

and Considering Damping Values for Propagation of P-Waves to be 1/3 Damping Values for Propagation of S-Waves  :

- July

f I

~

Frequency - HZ M!l-Q  : P9  ;

O.1 1 10 100 i

2. 5 i iiiiii i iiiin i All Cases Considered in CESSAR -

Case B-3.5 using .

i Poisson's Ratio = 0.30  ;

~

\ \ I 20 _

C _

,9 ~

1 g

.? 1. 5 O -

U .

T

~

% ~

b 1.0

. k ( l, q o iY, CO _

). '

\ y f

3O ~

a i

f' l

!c, ,

g 4 i k O. 5 , n -

(f ,( y' j ,.

. j ,

+

'7 -

~

/

f.'@; ' Damping = 0.05

~

~ "

O.0 Fig. 2-7 Comparison of Vertical Spectrum Calculated at the Foundation Level for Case B-3.5 Using a Poisson's Ratio ~= 0.30 with Spectra Obtained for all Cases in CESSAR Note: damping values for propagation of P-Waves were considered = 1/3 the damping values for propagation of S-Waves for all the cases shown in this figure.

July j t

2

Frequency - Hz w a as: P. I o 10 100 0.1 1 i i i iiin i

,,,,,o i 2.5 '

~ All Cases Considered in CESSAR Case B-3.5 using _

Poisson's Ratio = 0.47 I ' ' ' '

2.0 CD t

C ~

.e

~

i l

5e 1. 5 _

y _

O q _

g _

h -

a l O . /V a l CD (;

Q, l=0 nl 16 g

a

]l\ i e 4

Ca, a f .

h'/

~

% ~

J 3

i 2

f <

l t _

^

l} - , n' ,

0. 5 4 '
r. jft ,

h bN l iiisi 4

- l f ,

n -

,m' . Damping = 0.05

0. 0 .'

Fig. 2-8 Comparison of Vertical Spectrum Calculated at the Foundation Level for Case B-3.5 Using a Poisson's Ratio = 0.47 with Spectra Obtained for all Cases in CESSAR Note: damping values for propagation of P-Waves were considered = 1/3 the damping values for props:stion of S Waves for all the cases shcwn in this 1%ure.

July

~~-- ---- _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ "-

4 Peak Vertical Acceleration - g 0.2 0.3 0.4 0. 5 0. 6 0.0 0.1 ' '

0 o e

)

't' 20 N

u  ;

q -

b '

. 4g e t  :

b g

8 ..

e

$ 60 , ,','

5 ,

, Se - ,

ca ,

S

  • Case B-3.5 0 80 I Poisson's Ratio = 0.40 Q , f 'e $,

' --- Poisson's Ratio = 0.30 .

e

  • Poisson's Ratio = 0.47

,', N

?

100 m

Fig. 2-9 Calculated Peak Vertical Accelerations for Case 8-3.'5 Using Various Poisson's =

Ratios and Considering Damping Values for Propagation of P-Waves to be 1/3 Damping Values for Propagation of S-Waves for Case B-3.5 with Poisson's Ratio = 0.4 (Original CESSAR Case) and Damping Values for Propagation of P-Waves to be Equal to those

,,.r r~ c wmroe fnr rho Other Two Cases (B-3.5 with Poisson's ratio = 0.30 & 0.47)

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

E ENeL A8: P t2 Frequency - Hz O.1 1 10 100 .

2.5 ... . . . . ..... ,

- Ali Cases Considered in CEsSAR

Case B-3.5 using Poisson's Ratio = 0.30
2. 0

~

i -

i! -

q ), -

i t

w k [ l ,

.e 1. 5 i f,

O 1 i o

o ) t l

K ~

l v

i

\

s -

j j

[0[)N o 3

1. 0 M _

[ ,) _

'5 i ) i 1 )

j

~

.Q  !

l f

h 5 $ ;V dk [

0. 5 l "' V' 'b~ '
)

~

^}f D,

i

. l

/ i J ..

~

~ / O Damping = 0.05 ~

'I -- '

O. 0 .

Fig. 2-10 Comparison of Vertical Spectrum Calculated at the Ground ,

Surface for Case B-3.5 Using a Poisson's Ratio = 0.30 and Damping  !

Values for Propagation of P-Waves Equal to Those for S-Waves with Spectra Obtained for all Cases in CESSAR.

~

Note: damping values for prcpagation of P-Waves were considered = 1/3 the damping values for propagation of S-Waves for off the CESSAR Cases shown in this figure. _

I

m

"**: p 13  ;

Frequency - Hz ,

. O.1 1 10 100 4 2.5 .., , . . ...... . .

- All Cases Considered l in CESSAR Case B-3.5 using

~

Poisson's Ratio = o.47  :

A0 \

CD -

C -

1 -

.e m .

~

l 2 / g 9 1. 5 j i I 4

i D 1 -

O -

O  !

/ t h j -

t

, b')

l I

( l

} -

f -

) ,

) \ \

D b i \ ll, ' & . \ \

q 1. 0  ;

y n .

.g to ( . I, y

f \

'iB f y

g

)

p/ pypq \. '

q4.\ 4

~

J \

0. 5 hvh jij ;jU '< '

/l

.\

[Q

_ w -

-. 6 Damping = o.o5 -

d

0. 0 Fig. 2- 11 Comparison of Vertical Spectrum Calculated at the Ground Surface for Case B-3.5 Using a Poisson's Ratio = 0.47 and Damping Values for Propagation of P-Waves Equal to Those for S-Waves l with Spectra Obtained for all Cases in CESSAR. l 4

Note: damping values for propagation of P Waves were considered = 1/3 the demping values for propagation of S-Waves for all the CESSAR Cases shown in this figure. ,,

l

Frequency - Hz

" ' * * ' P 14 .l 0.1 1 10 100 2.5 iiiiin i i i iiiii, ,

All Cases Considered in CESSAR -

i Case B-3.5 using .

Poisson's Ratio = 0.30 _

l Il l' i 2.0 -

CD _

B C ~

.Q 2 ,

5 e

1.5  :

o -

o -

q _

% ~

~

8 .

o

  • e

. A .

jf Q *

%(j D

}h d T

[

3 I f J

"h a q ,

.e ^

f I,d2 O.5 ym '

p p% ' ,

b.

~

j .

I \>

Q .

j g. .- .

C'~ i i i l i s tii

/!, '

Damping = 0.05

(

l rh

O O. O Fig. 2-12 Comparison of Vertical Spectrum Calculated at the Foundation Level for Case B-3.5 Using a Poisson's Ratio.= 0.30 and Damping Values for Propagation of P-Waves Equal to Those for S-Waves with Spectra Obtained for all Cases in CESSAR.

Note: damping values for propsgation of P-Waves were considered = U3 the damping values forpropagation of S-Waves for all the CESSAR Cases shown in this figure.

l

Frequency - Hz

^

O.1 1 10 100

2. 5 i iiin, i i iiiiin i All Cases Considered

~ ~

. in CESSAR -

. Case B-3.5 using -

Poisson's Ratio = 0.47 _

2. 0 C -

.e 5 1. 5 e

e . -

O g _ -

, t -

$ 1.0  ?

O. \ '

~

,. [ ,

~

.? > f\ '\

a5 V;)w%e;{ ,

l

't k[ g af./ 3 f mM _

tiiiii 4

-. f. '

Damping = 0.05 -

0. 0 l Fig. 2-13 Comparison of Vertical Spectrum Calculated at the \

Foundation Level for Case B-3.5 Using a Poisson's Ratio = 0.47

  • and Damping Values for Propagation of P-Waves Equal to Those for S-Waves with Spectra Obtained for all Cases in CESSAR.

Note: dempir'g values forpropagation of P-Waves were considered = 1/3 the comping values for propogstion of S-Waves for all the CESSAR Cases shown in this figure.

EWL 38- Pn CONCLUSION

1. Vertical results are sensitive to variation in poisson's ratio
2. System 80+ vertical motion is high due to low damping used in the convolution analysis of vertical waves (damping was taken as 1/3 '

of that for S-Waves)

3. Current thinking of the industry is that P wave damping should be equal to S wave damping
4. Regardless of damping and poisson's ratio, site is qualified as long as vertical motion at the surface as developed by COL applicant for site specific conditions is under the envelope of System 80+

surface spectra

~

t ABB COMBUSTION ENGINEERING SYSTEM 80+  !

T. &lCL LB. : P - t ~7 l i

l

_. CUT-OFF FREQUENCIES USED IN SHAKE ANALYSES l June 8-10 Audit Item: (19) l QUESTION: )

The cut-offfrequencies used in the SHAKE computations impact the strain computations. For soft sites with lowfrequency cut-off, what was i

the impact offrequency cut-offon the highfrequency computation in the l soil column? What was the criteriafor selecting the cut-offfrequency? {

{

1 i

RESOLUTION:

1. A cut-off frequency of 40 Hz. was used in all SHAKE analyses. l l
2. The criteria was to choose the cut.off frequency high enough within l the frequency range ofinterest and consistent with frequency content of  !

the input motions. l l

l l

a  !

i l

(

4 I

I

{

(

l 1

ABB COMBUSTION ENGINEERING SYSTEM 80+

1

I 1

'1 EWL- cRB : p. pg DEEP STIFF SOIL SITES 1

June 8-10 Audit Item: (21)  ;

I QUESTION:

Are deep soil sites of relatively stiff soils covered in the range of columns investigated?

i RESOLUTION:

  • Two additional soil columns (cases E & F) were investigated ,
  • Case E was taken as 400' deep ,
  • Case F was taken as 800' deep
  • Soil stiffness variation with depth for both cases were agreed with the staff
  • Results attached for both horizontal and vertical analyses

! (

1 ABB COMBUSTION ENGINEERING SYSTEM 80+

r k

' #9 Shear Wave Velocity - fusec 0 500 1000 1500 2000 2500 3000 3500 0 . . - < ' '

50 j l

~

Case D-1 -

j Case E l 100 t _

e

$ 150 l M .

b c

l O ,

y 1 S S e l $ 250 '

n S ', Case D-1 g.

o l 300 ,

1 l Case E _

i 8

350  ;

I I -

^f ,

400 '

Fig. 4- 1 Shear Wave Velocities Used for CESSAR Case D-1 l and Shear Wave Velocities Selected for New Case E Jul:.

Frequency - Hz "" A * * " * * * -

O.1 1 10 100

2. 5

- All Cases Consideredin CESSAR -

_ Case E (400 ft Profile) .

2. 0 q _

.]

e k 1.5

?l}INy

' l 8

- \

I  ;

)1 l

~

~ a l \

\ l r.$ ya; r3 1. 0  ; yy '

t q ~

\';

% ^

)4 e E

'R i <ln ]'. 3

,k N

k h

N(I17

= A , iou, I

/,I l

,4 L i l i titil

, "Kl I I itIlli -

' ^

Damping = 0.05 f'

~

~

0. 0 Fig. 4-2 Comparison of Horizontal Spectra Calculated at the Ground Surface for Case E (400 ft Profile) with Spectra Obtained for all Cases in CESSAR Ju!

~

Frequency - Hz " 4 & : e 2I  ;

O.1 1 to 100 2.5 All Cases Consideredin CESSAR j Case E (400 ft Profile) _

~

Damping = 0.05 .

2. 0  ;

m _

i _ ,

t 9 _

e

&e 1. 5 U

U - ,

T _

'E

~

S 1.0

~

hM 1 aj M

h g

E -

hl%y

}

[

p p{'k

.8 -

v l1(

t H g o

1 y ,

% 1. _

'l . n M l

~

n

[ j

/ ) ) + -

.I s l /

'^ (

I,

-~

'! ll l l

0. 0

- l Fig. 4-3 Comparison of Horizontal Spectra Calculated at the Foundation Level for Case E (400 ft Profile) with Horizontal l Spectra Obtained for all Cases in CESSAR l 4

July

"" A B : P 12 Frequency - Hz 10 100 O.1 1 2.5 a All Cases Considered -

i in CESSAR Case E (400 ft Profile) ,

2. 0 -

D ,

t -

l q -

i _

9

~ _

b d

& 1. 5 ) ( -

j' b -

i l U - ,

) ai 0 -

l

% ~

j i I ,

E ~

~ I '

< >i "o [ M o i I I  !

5

1. 0 o

q ,

r,f

~

~

[ 'M I j ~

~

I "?

f $ .

i ML i_N

" yyr wl* .

)/

l

/ Damping = 0.05 -

0.0 l

Fig. 4-4 Comparison of Vertical Spectrum Calculated at the Ground Surface for Case E (400 ft Profile) with Spectra l

' Obtained for all Cases in CESSAR Note: Camping valves for propagation of P-Waves were considered = 1/3 the temping values for propagation of S-Waves for sit the cases shcwn in this figure. Jul)

EWL(8 : P. 23  :

Frequency - Hz  !

t i

O.1 1 10 100 j 2.5 i

All Cases Considered -

i in CESSAR _

t Case E (400 ft Profile) _ l

~

Damping = 0.05 2.0 j m y s'

C ~

~

9 ~

j 8  ;

1. 5 J2  ;

e ,

o l O .

q _

s -

u

~

y -

e

! 1.0 ( I

ro$ ,

,{- ( 4 f _

3 I I l h]h

. ~

y

~

s l

u - .

i, N -

{

i

'i M. I d$ l h;a 8

. \b M@QQlp .

f(. 5

t .

s i _

i , !'

~

'I

. 0. 0 4

W

~

Fig. 4-5 Comparison of Vertical Spectrum Calculated at the l

Foundation Level for Case E (400 ft Profile) with Spectra

. Obtained for all Cases in CESSAR l

l Note: damping velves for propagation of P-Waves were considered = 1/3 the damping values for pecpagation of S-Waves for all the cases shown in this figure.

l

? ~

.5 Lily ]

1

Shear Wave Volocity - ft/sec O 500 1000 1500 2000 2500 3000 3500 0 . . . - - . .

100 ,

200 t _

i o

O aw 300 - i t

D M _

_ i t

E '

@ 400 O  !

g So  :

500 S -

o  !

. 600 5

700 800.

l i

Fig. 4-6 Shear Wave Velocities Assigned to Case i~ \

t i

i 6

July . l l

i

~ v a , - - y

Frequency - Hz '^' " M : P 15 O.1 1 10 100  :

2.5

- All Cases Consideredin CESSAR .

- Case F (800 ft Profile) .

~

l l Damping = 0.05

2. 0  ;

O> .

~

y .

l e .

&o 1. 5

\

8 .

1()jl h -

q ,q e i ~

~

.\  ;

~ d f i \

/ %d l

{t f 'i S

1

@ 1.0  ;

M it c' 9 i /h f l

s .

! !y  : ^( \ \#9 / 1'N  :

/\j q,3' o

(. \A s

W

% f A \ i nio i

/l /

yJ j kl i I I illi ,

0. 0 Fig. 4-7 Comparison of Horizontal Spectra Calculated af the Ground Surface for Case F (800 ft Profile) with Spectra '

Obtained for all Cases in CESSAR July i

J Frequency - Hz 0.1 1 10 100 2.5 All Cases Consideredin CESSAR .

Case F (800 ft Profile) .

Damping = 0.05

. 1

2. 0 en .

g Q -

Se 1. 5 O -

Q .

Is .

l

~

i k ik r>  !

e 1.0 g s _ of1 h -

f y(

4 IJ j?

g l i

.N -

>l t c ~

, r v -

k

~"

j . b .

I (h \ i 7,P y_ -

/ IV/# %4 4m -

p' Q

~ / - i

s l

/ i O. O Fig. 4-8 Comparison of Horizontal Spectra Calculated at the j Foundation Level for Case F (800 ft Profile) with Horizontal  ;

Spectra Obtained for all Cases in CESSAR  :

l Jaly 1

~

J f

?

Frequency - Hz '" ' A 8 : P 1.-]

I O.1 1 10 100 \

I 2.5 .

-' \

All Cases Consideredin CESSAR Case F (800 ft Profile) l

--- RG1.60 Spectrum @ 0.3 g -

j

- 1 Damping = 0.05 ,

2.0 _

Ch .

e .

3 .

e .

R n t o 1.5 1i Y. by l

0 -

~ <

}

h U

\ \

$q ' '

~

(

~

/ -

m Q . i- {eM -}i!

l' l. .\

jj d{i( 11'

'O '

a h . \f' J'\p f \ .

l

! 'y l

k o3

~

I [h/ m A \ Td% k w ii ' ,

l i

9

\

w%"

.nnn e

i t L11Lut___ .

/)

,. f J , ~ iiiini  :

- l lllllll

[

\

' *- .4 / .J

0. 0 Fig. 4-9 Comparison of Horizontal Spectra Calculated at the Ground Surface for Case F (800 ft Profile) with Spectra Obtained ,

for all Cases in CESSAR and with RG 1.60 Spectrum .

July

frequency - Hz "" %8 : P 'E O.1 1 10 100

2. 5

' All Cases Considered in CESSAR Case F (800 ft Profile]

2. 0

~

J t ~

i

.Q -

%' /I l I l I

&e 1. 5 1 i

i

[s -

o -

I $

O 1 I d -

q -

/'

q

~ yI _

S I;N

\ -

- . l l /

e.

15 8 1. 0 l

- Jfh 4 w .

t \ \

%u .

t

.(

t ~

{

~

h _

3 C{ ('N_

!^ 'I O. 5 y '

m  ;

i

/,7 w -

f; f

l

\

~

r/ Damping = 0.05 -

'!'d 0.0 Fig. 4-10 Comparison of Vertical Spectrum Calculated af the ~

Ground Surface for Case F (800 ft Profile) with Spectra Obtained for all Cases in CESSAR Note: comping values for propagation of P-Waves were cons _idered = 1/3 the damping values for peccapstiour of S-Waves for s!! the esses shown in this figure. July

Frequency - Hz ~ A" ' "'

  • I O.1 1 10 100
2. 5 All Cases Considered -

in CESSAR _

Case F (800 ft Profile) _

Damping = 0.05

.2. 0 t -

.e

~

E S 1. 5 8 .

O q -

~ .

E -

Is  !

e . k . 4

%<g a _

f3 <l

~ .

,8 -

/ h,.;'

J E

0. 5 -

hNk1hk.3, --

I a \

J.j.

i s h _

. N .

s .

'I

0. 0 =

Fig. 4- 11 Comparison of Vertical Spectrum Calculated ai the Foundation Level for Case F (800 ft Profile) with Spectra Obtained for all Cases in CESSAR Note: Camping volves for propagatiorr of F Waves were considered = 1/3 the darnping values for propopstion of S-Waves for ett the esses shown in this figure. -

July x

Frequency - Hz "^'" 4 8 : P 3a 0.1 1 10 100

2. 5 ,.,, , , , , , , , , , . ,

~

i All Cases Considered in CESSAR ,

Case F (800 ft Profile) '

l

' - - - RG 1.60 Spectrum @ O. 3 g  !

2.0 - -

~

4 ,

I -

1 C -

9 ~

~

@ / l

.$ 1. 5 1 l, ca _

o -

I U / t K -

l \ y} ,. .I

';i ,

j}

, ~

% ~

k k

f -

13 e

a i dbl %j \, \

q 1. 0  ; o y ,

~

I;

/j

, ,l

.a I,

. ') .

E s , \j -

\_

0J 3

0. 5

.IJh/ n *,y I l .

,x p g. 4

- /- , x ,-

s l -

f

  1. :7 _

- j'l 1 Damping = 0.05

~

~~ *l l

~

.a O.0 Fig. 4-12 Comparison of Vertical Spectrum Calculated at the Ground Surface for Case F (800 ft Profile) with Spectra Obtained for all Cases in CESSAR and with RG 1.60 Spectrum fv'ote: darr. ping va!ues for propagation of P-Waves were considered = 1/3 the damping values for propagat>on of 5-Waves for el! the cases shown in this fgure.

~ July 1

(. EM eL QS, - P. S i CONCLUSION: ,

I

1. The surface motion from both cases E & F fall within the envelope of surface spectra used as the design basis for System 80+
2. System 80+ design covers deep sites -

1 l

l l

l l

I l

ABB COMBUSTION ENGINEERING SYSTEM 80+

ErJeLQS: P 32 WINKLER SPRINGS June 8-10 Audit item: (15)

QUESTION:

CE needs to show that Winkler Soil Spring and the uniform soil spring derivedfrom afinite element analysis are comatiblefor a governing SSI case.

RESOLUTION: -

  • The Winkler sensitivity model was re analysed with applied loads from case B-1 (CMS 1 motion) of the SSI analysis
  • Results are attached ABB COMBUSTION ENGLNEERING SYSTEM 80+

g .a, _ . - -& . - -u s a m . - = -

3  %

4 I

's, 9P

% g l %

. . i dh* I g ab,%  %

4 1P be ~,

6 g1P 'g 5 ,,  %

6 -

',, I'%

8's  :

9 I*g  !

%( l I's

%g% i 4*g a t 8% g db,% l 9%

db* , O t Ab% ,,. l d6% ,

e% g db% , O I%,

d6% , t .

db g I l

Jb% ,%,

46%  % .

g I  %

%q, 4 9%

%( l I

%, I d6%

g db t

i

n P

k=

, MOMENT (COUPLE) LOADING ALONG NORTH-SOUTH CENTERLINE STRIP ..

W

.v-, , , . , ...+, e-,w.- ..m, . . .,w...-..-..., ,,~~.,m,..

...,y..-.,4,,.., .m.-,w- o.,.- sr. -, - . , - - - . - < . , . . .---s. .. --w.--,u, - . . . . . , + , - - . - - .._e. . ,. . .- - - . . .._. , . . - - - . - .

I VERTICAL DISPLACEMENT OF BASEMAT DUE TO SPLli MOMENT - CASE B1/ CMS 1 0.003 - '-

- - - SoilSprings 0.002 -- _

Finite Element 0.001 '. -

s c -

C . ,

h

!s s ,  !,

E O . , , , , , , , , , , , , , ,

O 40 25 80.5 120.75 161 201.25 241.5 2'1.75 '. 322

_a 5 ,

3 -0.001 -- -

.o -

i t:

-0.002 - - ,

m

-0.003 - -

k t-

.Sd ta

-0.004 -

Distanco Along Basemol (!!) w

-k v - ,, ,-- ----e .ns- ,v w - , e---- - - , ,- n s , - - - - ,----r- - -

- ~ - - - - - - - - - - - - - - - - - - - -

MOMENT IN BASEMAT DUE TO SPLIT MOMENT LOAD - CASE B1/ CMS 1 6.00E+05 --


SollSprings Finite Element .

4.00E405 -- ,

2.00E405 -

c 8 '

6 -

. . . l . . .  ! . l

....[

. . . .-l = . . { ., , i I. I. . , ,

.} U.00C+00 _

0 40.25 80.5 120.75 161 201.25 .5 281.75 322 o

2 -

E -2.00E + 05 --

U _

-4.00E+05 --

. ft)

-6.00E405 --

s' kP kal 00

. -8.00E+ 05 --

1' Disionce Along Basemat (ft)

W

PRESSURE UNDER BASEMAT DUE TO UNIFORM SPLIT MOMENT - CASE B1/ CMS 1 8000 --

- t g_ 1

. - - - - - Soil Springs Finite Element 4000 --

2000 - -

e - .

d _

0  ! .' . j l , , , . . . . . ,

l . . . l . . . . . s , I g - .

2 0 05 80.5 120.75 161 .

201.25 241.5 2 1.75 ', 322 2000 --

4000 -- .

m

t

' '- P

.319 Go

,-8000 I Distance Along Basemat (ff) vi e

a m-- _ ... . -%, e . - r, -, . - - + - - , , .- - - - - - . - - - , , - - + - - - + - - - , - - - . . - - , , - - - + - - --.,c ,-w,e--- . - , , - , - . - - . -

N EWL- S S : A 17 i CONCLUSION:

1. Winkler representation is adequate for the purposes of static analysis of the Superstructure and the Basemat i

m.

ABB COMBUSTION ENGINEERING SYSTEM 80+

- -_ _ __ J