ML20206L894

Rev 1 to Review of Effect of 860131 Earthquake on Perry Nuclear Power Plant

ML20206L894
Person / Time
Site:	Perry
Issue date:	08/31/1986
From:	Jerrica Johnson, Mraz M, Robert Murray EQE, INC., LAWRENCE LIVERMORE NATIONAL LABORATORY
To:	Office of Nuclear Reactor Regulation
Shared Package
ML20206L865	List: ML20206L859 ML20206L870 ML20206L894
References
8666-02-01, 8666-02-01-R01, 8666-2-1, 8666-2-1-R-1, 8666-2-1-R1, NUDOCS 8608200395
Download: ML20206L894 (91)

Text

.

p s

• 8666-02-01 Revision 1 REVIEW OF THE EFFECT OF THE JANUARY 31,1986 EARTHQUAKE ON THE PERRY NUCLEAR POWER PLANT August 1986 l

Prepared for:

US NRC ,

Division of Boiling Water Reactor Licensing Office of Nuclear Reactor Regulation US Nuclear Regulatory Commission Bethesda, Maryland 20014 l

Prepared by: -

J. J. Johnson M.J.Mraz O. R. Maslenikov EQE Incorporated Two Annabel Lane, Suite 101 San Ramon, CA 94583 R. C. Murray H. J. Weaver Lawrence Livermore National Laboratory P O Box 808 Livermore, CA 94550 8600000395 PDR ADOCK $ $$ h 40 PDR E

V

t CONTENTS -

f.AS.t

1. INTRODUCTION 1-1 1.1 Background 1-1 1.2 Objective and Scope 1-2

2. REVIEW AND ASSESSMENT OF CEI TECHNICAL INVESTIGATIONS 2-1

3. SOIL-STRUCTURE INTERACTION AND STRUCTURE RESPONSE 3-1 3.1 General 3-1 3.2 Response Comparison-Vertically 3-5 -

i Incident Waves

' 3.3 Response Comparison-Non-Vertically 3-10 Incident Waves 3.4 Response Comparison--Structural Damping 3-16 4.

CHARACTERISTICS OF THE JANUARY 31, 1986 4-1 EARTHQUAKE MOTION 4.1 General 4-1 4.2 Energy, Power, and Duration 4-3 4.3 Response Spectra and Power Spectral 4-28 Density Functions 4.4 Nonlinear Response of Single Degree-of-Freedom Systems 4-51

5.

SUMMARY

AND CONCLUSIONS i 5-1

6. REFERENCES 6-1 l

l 1

• 4 CONTENTS (CONTINUED)

TABLES EASA 3-1 Summary of Soil Properties .

3-12 4-1 Comparison of Duration Measures of Selected 4-7 Acceleration Time Histories 4-2 Comparison of Energy, Power and RMS Acceleration 4-8 of Selected Acceleration Time Histories 4-3 Comparison of Duration Measures, Peak Ground 4-9 Acceleration, Peak Ground Velocity, Energy, Average Power and RMS Acceleration of Selected Input Accelerations

• 4-4 Scale Factors to Achieve Ductility Ratios of 1.85 and Corresponding Effective Peak 4-10 Accelerations Histories for Selected Acceleration Time FIGURES 1-1 4

Acceleration Response Spectra, Perry (1986) Reactor 1-4 Building Foundation and Containment Vessel 1-2 Auxiliary Building Recorded Acceleration Response 1-5 Spectra ~

3-1 Analytical Model of the Perry Reactor Building 3-4 by CEI (Ref. 2)

3-2 I Comparison of Calculated and Measured Response 3-7 on the Contai.. ment Vessel - Vertically Propagating Waves 3-3 2 Recorded on the Containment Acceleration Vessel Time Histories (ft/sec ) 3-8 3-4 2 Calculated Acceleration on the Containment Vessel Time Histories (ft/sec ) 39 l

11

3-5 Foundation Input Motions Calculated for Different 3-13 Angles of Incidence 3-6 Calculated Response on the Containment Shell for 3-15 Different Angles of Incidence 3-7 Comparisons of Calculated Response from Fixed-Base 3-17 Analyses with Recordel Response at El. 686 ft.

on Containment,Shell 4-1 AccelerationTimeHistory,PerryRgaetorBuilding Foundation (Perry, 1986), (ft/sec )

(a) North-South Component 4-10 (b) East-West Component 4-11 (c) Vertical Component 4-12 4-2 ' Acceleration Time History, Mitq' hell Lake Road New Brunswick (1982), (ft/sec )

(a) 180 from North 4-13 (b) 2880 from North 4-14 (c) Vertical 4-15 4-3 AccelerationTimeHistgry,PerryFoundation .

Design Motion (ft/sec )

(a) North-South 4-16 (b) East-West 4-17

. (c) Vertical- 4-18

• 4-4 Cumulative Energy (Equ. 4 -' 1) with Time, Perry I

Reactor Building Foundatien (Perry, 1986), (ft2 /sec3)

(a) North-South 4-19 (b) East-West 4-20 (c) Vertical 4-21 4-5 Cumulative Energy (Equ. 4 - 1) with Tina, Mitg/sec)

(ft 0 3 hell Lake Road, New Brunswick (1982),

(a) 18 0from North .

4-22 (b) 288 fecm North 4-23 (c) Vertical 4-24 4-6 Cumulative Energy (Equ. 4 - 1) with Time Perry Foundation Design Motion (ft2 /sec3 )

(a) North-South 4-25 (b) East-West 4-26 (c) Vertical 4-27 4-7 Acceleration Response Spectra, Perry (1986) 4-30 Reactor Building Foundation Motion 4-8 Acceleration Response Spectra, Mitchell Lake Road 4-31 iii

4-9 Acceleration Response Spectra, Perry 4-32 Design Reactor Building Foundation Motion 1 l

4-10 Power Spectral Density Function, Perry (1986)  !

(a) North-South 4-33  !

(b) East-West 4-34 l (c) Vertical 4-35 l

4-11 Power Spectral Density Function, Perry (1986) 1 Normalized to 1. g versus NRC Proposed Target (a) North-South 4-36 (b) East-Wett 4-37 (c) Vertical 4-38 4-12 Power Spectral Density Function, Mitchell Lake Road (a) 180 from North 4-39 (b) 2880 from North 4-40 (c) Vertical 4-41 4-13 Power Spectral Density Function, Mitchell Lake Road Normalized to 1 g versus NRC Proposed Target (a) North 4-42 (b) West 4-43 (c) Vertical 4-44 4-14 Power Spectral Density Function, Perry Design (a) North-South 4-45 (b) East-West 4-46 (c) Vertical , 4-47 4-15 Power Spectral Density Function, Perry Design Normalized to 1. g versus NRC Proposed Target

' (a) North-South 4-48 (b) East-West 4-49 (c) Vertical' 4-50 1

, iv

5 J

1. INTRODUCTION 1.1 Backaround At approximately 11:47 a.m. EST on January 31, 1986, an earthquake occurred of magnitude 4.9 Mb . Its epicenter was 11 miles south of the Perry Nuclear Power Plant. The earthquake was felt at Perry and motions were recorded at several locations in the Perry structures.

Three different types of instruments recorded the event at Perry. One type of instrument is the Kinemetrics Model SMA-3 strong motion time history recording accelerograph. This instrument records the three orthogonal components of acceleration over the duratidn of the-earthquake. Two of these instruments were installed-st Perry -- one on the reactor building foundation mat at approximately elevation 575' and the other is mounted on the containment shell at approximately elevation 686'.

Plots of the recorded acceleration time historf'es and calculated response spectra are shown in Fig. 1.1. These records best describe the characteristics of the earthquake ground motion, i.e, short strong motion duration (less than 1 sec.) and high frequency motion. .

1 The second type of instrument is the Engdahl PSR 1200-H/V response spectrum recorder. This instrument records the response of a series of single degree-of-freedom oscillators to the motion. It generates response spectral ordinates at discrete frequencies, in this case, twelve discrete frequencies ranging from 2 Hz. to 25.4 Hz., for a fixed damping value of 2% of critical. This instrument measures response i- spectra in three orthogonal directions. Four instrur$ ants of this type were used -- two on the auxiliary building foundation mat at an approximate elevation of 568', one on the reactor building foundation mat at an app'roximate elevation of 575', and one in the. reactor building

' on the drywell platform at an approximate elevation of 630'. Response spectra on the auxiliary building foundation are shown in Fig. 1.2.

srlperry.rpt 1-1

a. ~

Only the North-South component is plotted because selected discrete frequency vaises were unreadable in the East-West and Vertical components which makes plotting difficult.

The third type of instrument is the Engdahl Par 400 peak accelerograph.

It records.three orthogonal components of peak local acceleration. Two instruments of this type were used and were located on the auxiliary building foundation mat at an approximate elevation of 568' and on the reactor recirculation pump at an approximate elevation of 605'. A third instrument was out of service.

4 j .

1.2 Obiectivf_and Scone EQE Incorporated of San Ramon, CA was retained to assist Lawrence Livermore Natler,ai Laboratory (LLNL) and the US Nuclear Regulatory Commission in evaluating the effect of this earthquake on the plant structures and the soil-structure interaction (SSI) and structure response aspects of the event. The scope of this evaluation was three-fold:

a Review submittals by Cleveland Electric Illuminating Co.

(CEI) and its representatives and provide comments to LLNL and the US NRC. Meet with CEI personnel, NRC personnel, USGS personnel, and ACRS to discuss these submittals and the results of independent evaluations performed by EQE and discussed in subsequent sections.

Perform a plant walkdown. A summary of this effort is contained in Sec. 2.

m Perform independent analyses of the Perry reactor building and compare calculated and measured response.

Section 3 discusses the results of this effort.

m Investigate the characteristics of the January 31, 1986 earthquake compared to other recorded motions and the Perry Safe Shutdown Earthquake (SSE). The Fourier srlperry.rpt 1-2

energy, strong motion duration, power spectral densities (PSDs), and effective peak acceleration of the motions were calculated. Section 4 presents the results. .

Section 5 presents conclusions of this evaluation.

t srlperry.rpt 1-3

8 8 x to a) North - South x to b) East - West s.c .:

j 2 . 5- -

.+ -

c c 22.0-- I y

.> - l ig

ll  :

3 . .- -

ll .) Ii I

ll E

.. . / b M 1.e--

ll t '

\

4,v l% ~

1- -

l\ ,

.C  : " " " : :::. . ~.- . .

" ; ; . -- .C  : ; " ; .... .

" ",;1 s5 1

to' so S

to a ,gt ,,o , ,,2 Frequency (Hz) Frequency (Hz)

I s x to c) Vertical 8

a.C Legend: ,

I Reactor 81de fnd

• • ' lI E1. 575 ft.

E Containment Yessel 1ll1 O E1. 688 ft. _ __ _ _ .

. e..

Notes:

e

"+~

il All acceleration response spectra -

l l\g values in units of f t/ (sms)

. 2- -

All acceleration response spectra at N- 2X desping.

.c "

1 ..0 1 8 15 10 10 10 Frequency (Hz)

Fig. 1.1: Acceleration Responsa Spectra. Perry (1986) Reactor pepit.2 RSPLT2 Vo7/05/85 Building Foundation and Containment Vessel

s c

0 X 10 30.C l's 5

25.0- -

l\

I\

I \

e0.w-g I \

2a I k I e 15.0-- 't e

I l O

o 10.0--

v 1 a '

• s , o..

i rw' /

.C ,,

to 10 10 10 Frequency (Hz) i' Legend: Notes:

RCIC pump base mat North - South Translation E1. 574 ft.

HPSC pump base mat _ .__ _ . All accelerations in units of Fc/ (sus) .

All spectra calculated at 2% camoing.

F1C. 1. 2: Auxiliary Building Recorded Acceleration Response Spectra f191.2 ASPLT2 V07/05/85

s .

2. REVIEW AND ASSESSNENT OF CEI TECHNICAL INVESTIGATIONS Since the occurrence of the earthquake on January 31, 1986, we have attended several meetings with the utility (CEI), its representatives, US NRC, USGS, and the ACRS.

In addition, a plant walkdown of the Perry Power Plant was conducted shortly after the earthquake. A summary of the meetings follows:

Location Data Comments Perry Power Plant Feb. 10-11 Plant Walkdown; open meeting Washington, D. C. Feb. 21 NRC Staff /CEI

' Washington, D. C. Mar. 12-13 ACRS/NRC Staff

' Washington, D. C. Apr. 30 '

NRC Staff /CEI Washington, D. C. June 11 NRC Staff /CEI The utility and its representatives provided technical information in Refs.1-7, 9-12, and 21 assessing the effect of the January 31, 1986 earthquake on the Perry Nuclear Power Plant structures and equipment.

Numerous technical issues were discussed'and Ref. 8 itemized outstanding questions and comments for CEI's response. Reference 9 responded to these questions with additional follow-up in Refs.10 and 11 and telephone conversations.

l Our review concentrated on issues related to soil-structure interaction j

(SSI), structure response, and the effect of the January 31, 1986 l earthquake on structure capacity.

l The basic purposes of CEI's evaluations were:

e Quantitative assessment of the seismic qualification of a '

r comprehensive sample of equipment types located at l

various elevations in the Perry Power Plant structures, l

srl/prry2 2-1

This assessment covers equipment qualified by testing and analysis.

m Perform an evaluation of high frequency, short duration earthquake motions with regard to energy content and potential safety significance for structures and equipment at Perry.

Those aspects of these evaluations which relate to SSI, structure  ;

response, and structure capacity were reviewed.

Re-Analysis of Perry Structures To evaluate the s'eismic qualification of equipment located throughout the Perry structures, the structures were re-analyzed to generate in-structure response spectra for comparison with qualification spectra or  !

for re-analysis purposes. For rock-founded structures, the recorded l reactor building foundation time histories were used as input. Fixed-base analyses were performed. Structure damping corresponding to US NRC Regulatory Guide 1.61 was used and justified based on the levels of predicted response in the containment vessel using these values (see Sec. 3 for current study analysis tesults). The in-structure spectra were smoothed and peak broadened.

For the diesel generator building which is founded on fill, soil-

structure interaction analyses were performed by the utility in an identical manner to those performed for design--a finite element

approach. For the present re-evaluation, the input motion was assumed to be the time histories recorded on the reactor building foundation, i.e., these motions were assumed to be the free-field ground motion.

They were applied assuming they existed in the soil column rather than on a hypothetical rock outcrop which is more appropriate. Due to the soil column characteristics, however, the effect of applying the motion within the soil column is likely to add conservatism to the calculated response. Results from two soil profiles were calculated and presented.

Reference 10 reported results using low strain soil properties, whereas, srl/prry2 2-2

• 1 Ref. 11 reported results using the iterated soil properties from the design SSI analysis. It is likely that the soil properties most

compatible with the recorded motions lie between the two sets and closer '

to the low strain values. The resulting in-structure response spectra j differ significantly in frequency content for the two cases. For the low strain shear modulus case, significant amplification near 20 Hz is predicted. Equipment residing in the diesel generator building was evaluated for both cases and margin was demonstrated in both cases (Refs. 10 and 11).

The approach for generating in-structure response spectra for equipment evaluation is acceptable.

Structure Loads CEI calculated elastic loads in the concrete shield building (Ref. 21) and the containment vessel (Refs. 5 and 10). The stress levels in the containment vessel at the critical section were very low. The force levels in the shield building were significantly less than the design loads. Hence, from an elastic analysis standpoint, the Perry (1986) motions induced very low levels of stress in the structures. Section 4 discusses the inelastic response capacity of the motions for two simple structural representations. The evaluation of load levels in the diesel generator building was not explicitly made since in-structure peak accelerations were below design levels and, hence, would induce lower forces.

.The response assessment and load levels determined by CEI are acceptable.

Evaluation of the Effects of Hiah Freauency. Short Duration Earthauakes CEI evaluated the potential effects of the Perry (1986) recorded motions (and simpler representations) on ductile structures and components. One purpose of this assessment was to investigate the effect of high 4

frequency, short duration motions on components whose elastic frequency is near 20 Hz, the frequency range for which the Perry (1986) recorded j srl/prry2 2-3 1

,. _ _, ,,, - ,,___n__., _,.,, _,v_, ,,_ ,_,,..g_ ,- _.,-+g, ,

, o is near 20 Hz, the frequency range for which the Perry (1986) recorded motions exceed the SSE design spectra. The basis of this study is similar to that reported in Sec. 4--a recognition that design criteria for structures and components based on elastic analysis results specifies allowable stresses greater than yield. The behavior of the structure and component during the design level event is expected to be in the nonlinear range locally. Hence, a measure of the severity of the excitation can be its ductility demand on a particular structure or component.

CEI performed two sets of analyses (Ref.11). One set of analyses with ADINA as described below and a second set comprised of a number of parametric studies but for simpler forms of the excitation. In the ADINA analyses, six single degree-of-freedom models were developed --one for each of the six components of Perry (1986) recorded motions. The approach was to construct the single degree-of-freedom models.with frequency corresponding to the frequency of maximum spectral acceleration in the record. The six frequencies were near 20 Hz. An elastic analysis was performed for each of the six models subjected to the SSE design time histories at the corresponding locations. The loads calculated from the design time histories were assumed to be at the allowable value. The yield level was then set at one-half this value--

which is representative of several ASME design cases. Nonlinear analyses were then performed on the models. Each model behaved in an l elastic' perfectly plastic. manner. Ductility demands were then calculated for the Perry (1986) records and the SSE design time histories and compared. The ductility demand for the SSE design time histories exceeded that for the Perry (1986) records. Hence, the SSE l

design time histories were judged to be more severe than the Perry (1986) records even though the elastic response spectral acceleration of the Perry (1986) records exceeded the design.

A second set of analyses was performed on a similar model but for a simpler idealization of the input motion which permitted numerous parametric studies to be performed. One objective of these analyses was i

srl/prry2 2-4 i

to assess the effect of a higher amplitude, longer duration high l

frequency event. The study showed the impact to be relatively small on l

nonlinear deformation--the measure considered. The effects of preload . i and other quantities were assessed.

i l

CEI's conclusion was that the effect of the Perry (1986) motions on the  ;

response of ductile structures and components is less than the SSE design motion. In addition, the SSE design motion was judged to be more demanding on ductile structures and components than a longer duration, higher amplitude high frequency earthquake.

Although we differ with some elements of the approach, we concur with the assessment of the low damage potential to ductile structures and components of the Perry (1585) records and similar high frequency, short duration earthquakes.

l l

l i

l srl/prry2 2-5 i

- . . - - . . . _ . ._-.---,--,_,___...___..,.-_,__.___.-,.--_.._..--__._,.,----------_.-m-, - -

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

1 1

3. S0IL-STRUCTURE INTERACT!0N AND STRUCTURE RESPONSE 3.1 General The characteristics of the recorded motions on the foundations of the Perry reactor building and auxiliary building were shown in Figs.1.1 and 1.2. They are judged to be similar in frequency content to the 4

free-field ground motion, which was not recorded. The phenomenon which could lead to different foundation motion compared to the free-field is l soil-structure interaction (SSI) and structure response. SSI can be i

conceptually separated into kinematic interaction and inertial

interaction. Kinematic interaction is the phenomenon associated with wave scattering at the interface between the soil and the

' foundation / structure. Kinematic interaction leads to a different effective excitation of the structure from the free-field motion except I

when the foundation / structure are surface-founded and the wave propagation mechanism of the free-field motion'is vertically incident waves. In all other cases, the foundation input motion differs from the

free-field ground motion. Inertial interaction is the portion of SSI associated with the dynamic response of the combined soil-structure system when subjected to the foundation input motion.

All category I structures except the diesel generator building and the off-gas building are founded on very stiff rock (Chagrin shale with a shear wave velocity of 4900 ft/sec) or fill concrete of similar shear wave velocity. Even though the reactor building and auxiliary building foundations are founded approximately 50 ft. below grade, they can be treated as surface-founded at this elevation from an SSI standpoint because little or no side soil provides constraint or excitation due to

the presence of adjacent structures. Hence, embedmont is not considered. The very high stiffness of the rock is generally thought to preclude significant inertial interaction effects. Also, comparing the recorded motions on the reactor building and auxiliary building i

i

srlprry3 3-1

i foundations, shows them to be very similar. If significant inertial interaction would have occurred, the two foundation motions should differ because the structures have vastly different dynamic characteristics. Hence, it is judged that inertial interaction was not an important phenomenon. Potential kinematic interaction effects due to wave passage, i.e., non-vertically incident waves, were investigated in the present study for the reactor building.

The seismic design analysis of the Perry category I structures involved developing mathematical models of their dynamic behavior and analyzing them for the design ground motion. The ability of these models to predict response from the January 31, 1986 earthquake was investigated.

Linear analysis methods, as described below, were used to calculate response on the containment vessel at the location of the recorded motion. Calculated and recorded motions were compared. The motions recorded on the reactor building foundation were the excitation.

Several analyses were performed to investigate the effects of various parameters on containment vessel response.

The analyses proceeded by obtaining CEI's reactor building dynamic model (Fig. 3.1) (Ref. 2) and implementing it on the LLNL computer system.

During this effort, the structure model was modified in three ways.

First, the soil springs located at the base of the model were deleted so that a fixed-base model was obtained. Second, the model of the upper portion of the containment. vessel was modified to treat the crane as it was positioned during the earthquake. The mass of node 11 (elev. 725')

was reduced and an additional node having the mass of the polar crane i

was offset 50 ft, south of the containment vessel centerline and rigidly

. linked to node 11. This modeled the eccentric mass effects of the crane as parked during the event. Dynamically, it couples torsion and east-west translation and increases rotational inertia. Third, a massless node was added at the actual location of the containment vessel instrument and rigidly linked to node 13 which allows one to include the effects of torsion and rocking on the predicted response.

srlprry3 3-2

. _ _ _ - _ - _ . _ , _ _ -. , ,.___ _._ _ _ -_- _,._._ _ _ , _ ,._ _ ...___ _ _-__._._....._ ~ . _ . . . _ ~ - . - . _ - . _ . _ . _ - .

. o All structure response calculations were performed with the computer program CLASSI (Ref. 12).

srlprry3 -

3-3

J e e .

- - k hhh 1. #

Ml4 I 'ii ai=l

-I 22 l

l~.  :: sii 1:

ls  ! 3 te! a l i it' :s la e i,  : 12 75 a -

13ST3A Wo10v38-

• h

}'*h$

L j {*g fcd

- 94 a 11Milm:.I ig 11YM 0131HS 7vossolois~

v % it

• i }

- s 58d 1 g1 l

l E 5 s d d d -

s I = .m

\ nN- .* E 1* a.ama - -

% y 4y ed I

U th 8

ed 3 3 5 l a 8 as e d d dd osmae v - = e@ g IN3nmv1NO3- g E gE;EE gEmE E e E I y f o 9 4 % . i il r 1 1 j 1 i l, i dd I

d d d d c d d d. e

.I c O

SN10r19 Ol3fMS- m o ^EM% E ^UoE

'y y

^ .E a.  %

E- -

'o *9 9 y  % e

)d .f C

I d

.) 'I d d

$ 3 d d d

)d, 3 d

wS O

u e4 0

• C4

'O u a O #c N OQ0; W

'* M>

' ou-

-4 W MCH

'3 s*

'3' ,

xmm j [

, u

/ s a:- ce>

4 4 .O

• ... s *

, *g 7 ,

f/ \  ;

t\ ) ,

/ -

m

' - _C_ )_

D- . i ,

D if e4 g '

A

= _ _v . - ,- ,

F 'ap srlprry3 3_4 l

.. s.

3.2 Resoonse Comoarison-Vertically Incident Waves Figure 3.2 compares 2% damped response spectra, calculated and measured, on the containment vessel at elevation 688'. Structural damping was a constant 4% for all modes. Vertically incident waves were the wave propagation mechanism. Structural damping and non-vertically incident waves were the ' subject of sensitivity studies reported in subsequent sections. The fully modified structural model was used in this analysis.

The response spectra comparison can be separated into three portions:

i low frequency response (less than 10 Hz), response near 20 Hz, and zero period acceleration (ZPA). For the N-S component, there is no specific low frequency peak. Near 20 Hz and at the ZPA, the responses are under-predicted by 20-30% by the analysis. For the E-W component, the measured response has a low frequency peak at about 4 Hz which is not reproduced in the analysis. The analysis amplifies motion near 7 Hz.

• Examination of the E-W modes in this frequency range and other sensitivity studies show both peaks to be a function of the input i

motion. Near 20 Hz and at the ZPA, the calculated responses over-I predict the measured values by about 35%. For the vertical component, the low frequency behavior matches well in amplitude and frequency content. Near 20 Hz and at the ZPA, the calculated responses over-predict the amplitudes of response by about 30%.

An inspection of the calculated and measured acceleration time histories can be done from Figs. 3.3 and 3.4 where the time histories are shown on j

an expanded time scale. (Note, Fig. 3.3 differs in scale from Fig. 3.4 and between Fig. 3.3 (a), (b), and (c)). The strong motion portion of the recorded time histories is greater than that of the calculated

[ motions. Also, a beating-type phenomenon is observed in the N-S and vertical components which led CEI to hypothesize that a portion of this motion was induced by secondary causes, such as the polar crane (Refs. 4 l

j srlprry3 3-5 l

J u and 11). This remains somewhat of an open issue relative to the predictive capability of the dynamic models. For evaluation purposes, CEI used the as recorded motion at elevation 688' to assess equipment qualification at this location.

i srlprry3 3-6

x 10 m

a) North - South x to

• b) East - West 3.C 0 2 . .- -

.e- -

l

$ 2.0- - I h i

= .. . i .. .

;I  ; g

< t . 0--

i  :

4 \

. 2- -

\ Y

.P Q _

.c . : : : : . . .. .  : : . . :.  : . : . . .: .0 . ;^..... .  : : :::::  : ..".

0 I I 0 I 2 16 10 10 10 to 10 10 10 Frequency (Hz) Frequency 04 1 w

.J I

x 10 c) Vertical Legend:

1. 0-. E1. 688.5 ft.

Calculated e E1. 686 ft. Measured o .

a E Notes:

e . 6--

j ,

All accelerations in units of f t/ (sms) y ,# .

I All spectra calculated at 2% damping.

. 2--

\w -

.c . . . . . . . . :.....  :  ::: :

10 to 10 10 Frequency 012)

Fig. 3. 2: Comparison of Calculated and Measured Response on the RSPt.T2 V07/05/es Containment Vessel - Vertically Propagating Waves fig 3.

,, e -

a*

f. '

d . . ..

v .

"' 2 itser femra (a) North-South m e. '

e...

y -e...

-e.e..

.. ..~. e.. s.. 4l. ,

T!stE (SEC1 (b) East-West E .* g w :'= m s M y 4.Qg -_ =

j-1 l ti e ese ci (c) Vertical Fig. 3.3 Recorded Acceleration Time Histories (f t/sec2) on the containment vessel.

3-8 l

I

> s io -

N o ,,,

^'

%AA ,ja,g 7 ,, * ;. , . . _ _ _ _ j ,AA 1 aAA .

s

-1o -

,IjDI,,,,A,,..,v_,,.

o o.s 1.o 1.s 2.o 2.s s.o s.s (a) North-South *** '*'

to -

o ^^j,h^ y _ ; ; , .-.

^ ,,,

l' ^ * *:,,,

a ^:fl1 p

-io -

o o.s i.o S.s 2.o s.o 2.s s.s (b) East-West Tmc (se to -

o ' 'j'

^- ^

. y ., ", ,8f '," ' ; c.,"".,^* - ""

_,,_  ; -f .g 8 *[fg a g, ..

8

-to Illl o o.s 1.o 1.s 2.o 2.s s.o s.s (c) Vertical Fig. 3.4 Calculated Acceleration Time Histories (ft/sec )

on the Containment Vessel 3-9

e .

3.3 Response Comoarison- Non-Vertically Incident Waves In an attempt to understand the causes of the discrepancies in measured l and calculated response discussed in Sec. 3.2, the effect on response of non-vertically incident waves was assessed. In particular, we sought to

, investigate the difference in amplitude of the near 20 Hz spectral accelerations and the low frequency behavior in the E-W component. We i considered these discrepancies to possibly be due to unrecorded foundation rocking and torsion. The effect on response of non-vertically propagating waves with their induced rocking and torsion was assessed. Two cases were considered--incoming waves from the south with varying angles of incidence and incoming waves from the east with varying angles of incidence. Only the former case is reported here.

The latter provided no new information.

4 The effect of non-vertically incident waves generally is to reduce the horizontal foundation motion in the direction of wave propagation,

, increase rocking in this direction and induce torsional motion. We j studied this effect on the response on the containment vessel for non-vertically incident waves of 10 0 , 300 , 45 0 , and 60 0 off the vertical axis and originating from the south. The method used was approximate in i that foundation input motions were calculated from the recorded 1 foundation translations assuming only their phase changed as they i traversed the foundation. The foundation was assumed to be located on

] the surface of a uniform half space having the properties of the Chagrin

shale layer directly beneath the reactor building foundation. Table 3.1 l summarizes these properties. The resulting foundation input motions were used in fixed-base analyses of the fully modified structural model j using 4% damping in the structure. Response spectra at 2% damping are shown in Figs. 3.5 and 3.6. Also included in these results are those j for the vertically propagating wave case. Figure 3.5 shows the

calculated foundation input motions. One can see from the plots that

! above 10 Hz the foundation translations decrease with increasing

} inclination angle. One can also see that rotations increase l

1 srl/ perry 3.3 3-10 i

-,--~,.-~.--.--c..m--,,,----.---,~..---,,-__.--.,,-v,-m,-,,.-.. _.,,.-r--,,--.,-,-,-w,-..,.,2--,y---,-em-,~~-,---.--,

n s

significantly with increasing angle. Figure 3.6 shows the calculated .

response on the containment vessel at the location of the instrument.

For all components the spectral peak near 20 Hz decreases with increasing inclination angle. Response in the N-S direction increases at 4 Hz with increasing angle, but E-W response is unaffected below about 7 Hz. Comparison of these spectra with those of the recorded motions shows little improvement in agreement for non-vertically incident waves. Thus, it does appear that vertically incident waves lead to the best estimate of containment vessel response; especially the near 20 Hz spectral acceleration.

a i

I srl/ perry 3.3 3-11 l

.. c Table 3.1 Summary of Soil Properties compression Wave Velocity 10400 ft/sec Shear Wave Velocity 4900 ft/sec Unit Weight 152 lbs/ft 3 Poisson's Ratio 0.36 Material Damping Ratio 0.01 I

e G

0 3-12

8 x to a) North - South X to b) East - West

.4 ts.c 16.o-- h ,

c

.y.

c 14.0--

h

)'

a \

2 12. &-

u of

? (\ f 10.o--  ?\ p e_ . 2-~ e, '\ l

. 8 . 0-- g U e- 0

< < s . o-- j15 8-- 4 . o--

w. 2...

.0 . . . ... .. .c -

16' 10 0

10 1

10 2 ggt 3,0 go t 3,2 Frequency (Hz) Frequency (Hz)

U Y x to c) Vertical

• w 30.0

• w Legend:

! 25.o-. Angle = 0 0 10 0 Angle = - _ _ _ _ _ _ _ _ _ _ . _ _ .

c i

2 20.0--

Angle = 30 0 _ _ . . _ _ . _ _ _ . _ _ _ _ _ _

' N- Angle = 45

" t s . &- )l4

,, Angle = 60 0 __ __ _.

I U

y so,o_. L Notes:

I All accelerations in units of f t/}sWs)

)k ,

5.&- D All spectra calculated at 2% damping.

.o ,

to 10 to 10 1 Frequency (Hz) i Fig. 3. 5: Foundation Input Motions Calculated for Different l RSPLT2 vo7/o5/85 Angles of Incidence - Translational Components ( 193. !

0 ~I x 10 a) Rocking about E-W axis x 10 e) Torsion about Vertical axis l 1. 4-- p 1

4--

e I**~

e t ,

2 > 2 1. 0--

'i a y. l'a a 4 e 6 e tJ 1

II I -

I I a

. e-. '9'l

- 11 3 .o- h E u . s-- e JI a i o

< 44

.+- /

.1--

• , Of f *o. J \ \g,h~~ s J I W.2 .:. p

~~

.c J n., 9

. --.?

.C . .

. _. . /. . .

..0 I 2 ggi 10 I

10 t6 I 10 10 10 10 Frequency (Hz) Frequency 0121 w

l

- Legend:

4 s I Angle = 0 Angle = 10* ______

0

! Angle - 30 _ _ _ _ _ _ _ _ _ . _ _

i Angle = 45 Angle = 60 .. .. _.

i Notes:

i

! All accelerations in unics of rad /(sus I All spectra calculated at 2% camping.

4 .

l Fig. 3.5: Foundation Input Motions Calculated for Different Angles of Incidence - Rotational Components r ig3.!

RSPLT2 V07/05/e5

i 2 a I x 10 a) North - South x to b) East - West

.c

1. +-

1. 2-- . 6--

J o o 01.0-- -

O.

L L

= . e-- . .+- 1

=

3 . s-- 8

< < q

.+- . 2--

. 2-- s

~m .o . .. .

.0 I

I 10 2 g3 1 10

.0 10

.I 10 16 10 10 Frequency (Hz) Frequency Dizl w

b en x 10

• c) Vertical Legend: ,

1. &- Angle = 0 o

. Angle = 10 _.__ _ _ _ _

e o

. . e-- Angle = 30 _______._____

o

a. Angle - 45 L o

. e-- Angle - 60 .. _. _.

U .

y .__

Notes:

1 All accelerations in units of f t/ (sWs)

> e- s All spectra calculated at 2% camping.

.c . . . . . . . . .: ,

10_ 10 10 10 Frequency (Hz)

I Fig. 3. 6: Calculated Response on the Containment Shell for RSPLT2 V07/05/85 Different Angles of Incidence f ig3.s

s .

I 3.4 Resoonse Comoarison--Structural Damoina Early in the present investigation, we studied the effect of structural damping on response of the containment vessel.

The only structural model modification incorporated in this

, study was considering a fixed-base structural model. Hence, the responses reported here are on the centerline of the containment vessel and do not include the crane eccentricity or a rigid link to the instrument location on the shall itself. Quantitatively, the effect of damping on response can be observed from Fig. 3.7. Of note, however, is that the 3% damped case does not increase containment vessel i

response near 20 Hz enough to match recorded motion in the N-S direction. In the E-W direction, the 5% damped case

.l still slightly over-predicts response. Finally, vertical response at the shell centerline is independent of structural damping. -comparing vertical response of Fig. 3.2 and Fig. 3.7 shows the importance of rocking at elevation 688' on the vertical component.

l The damping parameter variations investigated here did not produce changes in response large enough to permit predicted response to match calculated response near 20 Hz or at the

ZPA.

l 1

i t

i i

l 4

l 3-16

x to # South - Translation x go 2 West - Translation 3.C ,-

2 . 5-- g c **"

c -}

3 2.0- o,

" e

• a L i g a t.5- i . , 4_. 3

• 7o u

%t 0-- y 4

. 2--

6

. 5--

~_

! .0 . .

. :n .0 . w I I ..0 10 I

10 2

16 10 10 10 16 10 Frequency (Hz) Frequency (Hz) y x to # Vertical - Translati 3 w t.C 4 Legend:

I -El. 686 ft Measured

. e--

El. 688.5 ft E Calculated 3% structure damp y .3 _

4% structure damp _ _ _ _ _ _ _ _ _ _ _ .

5% structure camp .

u 4--

u

• Notes:

. 2-- All accelerations in units of ft/sec2.

All spectra calculated at 2% damping.

=

.0

..10 I 16 10 10 Frequency (Hz) i Fig. 3. 7 Comparisons of Calculated Response from Fixed-Base Analyses with Recorded Response at El. 686 ft. on Containment Shell q ,,

aset 2 s r 'as 'as

4. CHARACTERISTICS OF THE JANUARY 31, 1986 -

4 EARTHQUAKE MOTION 4.1 General There is a vast literature which documents the low damage potential of earthquakes of short duration and high frequencies. One of the most recent was that of Kennedy et al. (Ref.13). In their study, Kennedy et al. investigated several characteristics of recorded earthquake time j histories--Fourier energy, strong motion duration, root-mean-square (RMS) acceleration, response spectra, effective peak acceleration, etc.

Eleven recorded earthquake ground motions were investigated. The eleven were divided into two groups. Group I was comprised of records denoted Taft (1957), Olympia (1949), El Centro #12 (1979), El Centro #5 (1979),

1 Pacoisa Dam (1971), and Hollywood Storage PE Lot (1971). Group 2 was comprised of records denoted Coyote Lake (1979), Parkfield Cholame #2 (1966), Gavilan College (1974), Goleta (1978), and Melendy Ranch (1972).

The precise identification of location and components is contained in Ref. 13. The ~ earthquake records comprising Group I have strong motion durations greater than or equal to 3.4 seconds, are from earthquakes with local magnitudes of 6.4'and greater, and are rich in frequency content from at least 1.2 Hz to 5.5 Hz. The Group 2 records represented different earthquake characteristics. Their strong motion durations were less than or equal to 3 seconds. They are from earthquakes of local magnitude 5.7 or less. They have much narrower frequency content than the Group 1 records. One purpose of the Ref.13 study was to investigate differences in identifying characteristics and damage potential of the Groups I and 2 earthquake records. In addition, an artificial acceleration time history whose response spectra approximately matched the US NRC Regulatory Guide 1.60 design ground t response spectra was included. The total data set was 12 records.

The present investigation of the January 31, 1986 Ohio earthquake recorded motions proceeded in a similar fashion to Ref.13. The recorded acceleration time histories on the reactor building foundation srl/perryrpt 4-1

" ~

,1 were used and for brevity are denoted Perry (1986) earthquake motions.

Neither the Group 1 nor the Group 2 data sets match the characteristics of the Perry records which are richer in high frequencies (near 20 Hz) I and of shorter strong motion duration. The Gavilan College, Hollister, 1974 record and the Melendy Ranch Barn, Bear Valley, 1972 record of Group 2 are the most similar in terms of having a short strong motion duration and possessing somewhat higher frequency content (but still less than 10 Hz). The two horizontal components and the vertical component recorded at Mitchell Lake Road during the March 31, 1982 New Brunswick earthquake were also considered. Both sets of data are characterized by short, strong motion durations and rich high frequency content, i.e. greater than or equal to 20 Hz. In one sense, these records can be considered as comprising a Group 3. Finally, the Perry Power Plant design time histories calculated on the reactor building foundation were considered.

i srl/perryrpt 4-2

~

4.2 Enercy. Power. and Duration This investigation of the Perry (1986) records parallels that of Ref.

13. This permits one to make a direct comparison of the characteristics of the Perry (1986) records, the Mitchell Lake Road, New Brunswick (1982) records, and the Perry design motions with those of the Ref.13 data set.

Eneroy The total energy Em (or more properly the Fourier energy) of the records is defined as:

to+TD E(T) - a2 (t)dt (4-1) t '

o where E(T) is a measure of the total energy between timesot and to+

TD , a(t) is the instrumental acceleration at time t, and T D is the duration of strong motion. This approach corresponds to Ref. 13 and this definition of E(T) is attributed to Arias (Ref.14) and Housner r (Ref. 15). Cumulative energy plots as a function of time are shown later in this section.

Power The average rate of energy input (earthquake power) is given by:

E(T)

P -

(4-2)

TD and the root-mean-square (RMS) acceleration is defined to be:

ARMS - (4-3) srl/perryrpt 4-3

Duration Energy, power and the RMS acceleration are strongly influenced by the strong motion duration TD . In the past, many definitions of strong motion duration have been used. Two measures of strong motion duration are reported here -- two of several reported in Ref.13.

The most common definition of strong motion duration is due to Trifunac and Brady (Ref. 16). By this definition, the stronig motion duration is defined as:

T D - T0 .95 - T 0.05 (4-4) where T0 .95 represents the time at which E(T)/E, = 0.95 and T0.05 represents the time at which E(T)/E,= 0.05. This duration includes 90%

of the energy.

A second measure of strong motion duration is presented in Ref.13 and it is argued that it should be used when evaluating stiff structures such as those of nuclear power plants. The~ definition is:

T'O - Tg - T0.05 (4-5) where Tg - Max T0 .75 T

PA (4-6)

, s and T0 .75 represents the time at which E(T)/E,- 0.75. TPA represents the time associated with the first crossing of the accelerogram following the maximum positive or negative acceleration, whichever occurs later in time.

The power for this definition of strong motion duration (T'0), can be defined by:

aE P- (4-7)

T'D srl/perryrpt 4-4

where a E represents the cumulative energy between time T0.05 and Tg.

a E is equal to 70% of E,except when Ty exceeds T0.75 For that case, A E is greater than 70% of E,.

, Results Figures 4.1, 4.2, and 4.3 contain plots of the acceleration time histories for Perry (1986), Mitchell Lake Road, New Brunswick (1982),

and the Perry foundation design motions. Figures 4.4, 4.5, and 4.6 contain plots of cumulative energy as a function of time, again for the three sets of earthquake records, respectively. The energy, duration, power, and RMS accelerations for these records were calculated and are presented in Tables 4.1 and 4.2. For comparison purposes, Table 2-3 of Ref.13 has been reproduced as Table 4.3.

consider the time history plots of the Perry (1986) records shown in Fig. 4.1. Note, the two distinct phases of motion in the time histories. These two phases lead to estimates of strong motion duration TD and T'D which are slightly misleading. The initial phase of the -

motion contributes greater than 5% of the total energy; hence, estimates

, of strong motion duration begin in this phase and extend into the second phase when 75% or 95% of'the total energy is achieved. The strong motion duration estimates T'D for the Perry (1986) records are 2.38 4

1 sec., and 2.54 sec., and 2.40 sec. for the N-S, E-W, and vertical components.

The cumulative energy plots of Figs. 4.4, 4.5, and 4.6 show the concentration of energy over a short time period for the Perry (1986) records and the Mitchell Lake Road, New Brunswick (1982) records compared to the Perry foundation design records. In the latter case the rate of increase in energy over the time duration of the records is relatively uniform.

A comparison of the energy in the Perry (1986) records with the energy in the records: considered in Ref.13; the Mitchell Lake Road, New 4-5

Brunswick (1982) records; and the Perry foundation design motions shows the Perry (1986) records to have less energy than any of the records considered. Comparing the Perry (1986) records to the foundation design motions shows the Perry (1986) records to have 3.25%, 1.75%, and 1.94%

of the energy in the design motions for N-S, E-W, and vertical components. -

The Mitchell Lake Road, New Brunswick (1982) records were included here to provide another example of recorded motions with significant high frequency content. From Tables 4.1 and 4.2, one observes that estimates of strong motion duration for the three components are less than I sec.

for T'D and less than 1.5 sec for TD. The instrumental peak acceleration, the energy, and the RMS acceleration in the vertical component are substantially higher than those of the two horizontal components.

e a

i l

l srl/perryrpt 4-6

i Table 4.1 s Comparison of Duration Measures of Selected Acceleration Time Histories Earthquake Peak T pg TD T'D (sec )

i Record Accel T.05)

(sec T.75)

(sec T.95)

(sec (sec) (sec)

(g)

Perry, 1986 .18 0.24 2.62 5.85 2.55 5.61 2.38 (N - S)

Perry, 1986 .103 0.15 2.69 10.13 2.42 9.98 2.54 (E - W)

Perry, 1986 .105 0.47 2.87 5.15 2.61 4.68 2.40 (Vertical)

Mitchell Lake Rd .15 0.16 0.99 1.56 0.53 1.40 0.82 New Brunswick, 1982 (18 0 from N)

Mitchell Lake Rd .24 0.35 1.03 1.41 0.82 1.06 0.68 New Brunswick

• 1982 (288 0 from N)

Mitchell Lake Rd .58 0.10 0.78 1.23 0.35 1.13 0.68 New Brunswick.

1982 (Vert)

Perry Foundation .18 2.08 15.68 19.27 16.94 17.19 16.94 Design Motion (N - S)

Perry Foundation .17 2.32 16.06 19. 30 12.72 16.98 13.74 Design Motion (E - W)

Perry Foundation .19 2.90 15.52 19.07 10.20 16.17 12.62

Design Motion (Vertical)

I Table 4.2 Comparison of Energy, Power and RMS Acceleration of Selected Acceleration Time Histories Earthquake Peak T'D aE P RMS Accel Record 2 3 (g 2 x 10-3)

Accel (sec) (ft /sec ) (g)

(g)

Perry, 1986 .18 2.38 1.67 .678 .026 (N - S)

Perry, 1986 .103 2.54 .96 .366 .019 (E - W)

Perry, 1986 .105 2.40 1.10 .443 .021 (Vertical)

Mitchell Lake Rd .15 0.82 2.90 3.39- .058

^ New Brunswick E 1982 (18 0 from N)

Mitchell Lake Rd .24 0.68 5.26 7.46 .086 New Brunswick 1982 (2280 -from N) .

Mitchell Lake Rd .58 0.68 15.3 21.7 .147 New Brunswick 1982 (Vert)

Perry Foundation .18 16.94 51.45 3.65 .060 Design Motion (N - S) ,

j Perry Foundation .17 13.74 54.9 3.85 .062 Design Motion (E - W)

Perry Foundation .19 12.62 56.8 4.34 .066 Design Motion 4

(Vertical)

t Tablo 4.3 Comparison of Duration Measures, Peak Ground Acceleration, Peak Ground Velocity, Energy, Average Power and RMS Acceleration of Selected Input Accelerations (Table 2-3, Ref. 13) j  !

T T AE P RMS Acc.

O O a v Earthquake Record # 3 (g2x10-3)

(Component) (sec) (sec) (g) (in/sec) (ft /sec ) a m,(g) 1 Olympia, WA., 1949 6.7 64.2 3.97 .063 15.6 17.3 0.281 (N86E) 2 Taft, Kern Co., 1952 7.0 27.4 2.57 .051 10.3 28.1 0.180 (569E) 3 El Centro Array No. 12 6.9 18.6 1.88 .043 Imperial Valley, 1979,(140) 9.6 18.6 0.142

^ 4 Artificial 13.0 0.200 11.3 44.2 4.54 .067 E (R.G. 1.60) 9.4 5 Pacoima Dam 44.6 465.8 74.0 .272 San Fernando, 1971 (514W) 6.1 7.4 1.170 6 Hollywood Storage PE Lot, 0.211 8.3 30.0 5.37 .073 San Fernando, 1971 (N90E) 5.4 11.7 7 El Centro Array No. 5 17.3 78.1 22.2 .149 Imperial Valley, 1979 (140) 3.4 8.2 0.530 8 UCS8 Goleta 0.347 15.7 57.3 18.5 .136 Santa Barbara, 1978 (180) 3.0 9.7 9 4.0 13.3 5.86 .077 Gilroy(Array 1979, 050) No. 2. Coyote Lake,2.2 7.5 0.191 10 Cholame Array No. 2, Parkfield 10.4 86.1 59.4 .244 1.4 9.2 0.490 1966 (N65E) i 11 Gavilan College 1.6 2.1 1.80 .042 Hollister, 1974 (S67W) 1.1 1.6 0.138 12 Melendy Ranch Barn, Bear Valley 0.520 5.4 29.8 36.0 .190 0.8 2.6 i 1972 (N29W)

e 8

x to s . 0--

4. 0--

2. 0- -

z 1 L ai w A m._ ._...__m ._-.om. 2.. __ m m u.1 . a . _ _ .. -,

'"'r''''''''d 1

o .& F V w r ' r '" ' " r ' ' r ' - '"' ' ' ' ' ' "' '

4 I

h -2.0- -

_J

! W i o

! u o _4 . o- .

i 4

! w o

-e . o- - .

.o zlo 4.'o s .'o a.'o so.'o 12.'o 14.'o is.o ,

X 10 TIME (SEC)

Fig. 4.1 Acceleration Time History, Perry Reactor Building Foundation (Perry, 1986), (ft/sec 2) ,

(a) North-South Component a

! srl/prrytbl

1 8

x to 1

s . 0- -

2. 0- -

i

1. 0- -

_a LuJmL m .sJ.L ab g Jn. &. a.t. L. u m .i a . a i.a g ,, j.p vir 5-9 yvT -

gn p.g ,,,,,y'y'y..--. ,p r-,,.7

! g

H

! < - 1. 0- -

E w

w U -2. 0 -

U i

.m a  : -3.0-- . .

4 g . . .

~ ta.'o 14.'o ts.o 4.'o s.'o a.'o so.'o X 10 I

.o a.'o TIME (SEC)

Fig. 4.1 Acceleration Time History, Perry Reactor Buildin'g Foundation (Perry, 1986), (ft/sec 2) 1 (b) East-West Component .

l -

I i

'l T

S l

l o .

X 10 3 . 0- -

• i j

l t . 0- -

1 1

i 4

s . o--

I g *, .

L i 11 a n . u . . A 2 a 1. . m . _ . . . . - ,.a..m.-_m..,,

4 a e

F 7 1 r T =r T '8 7T v r ir-- r-- m - r-' - r - , -n- ir ' - ' r 4

- tr - 1. 0--

u w 1

w l

< u -2. 0- -

o

-3.0--

.o abo 4.'o abo a .'o sobo sa.'o 14.'o 1

X i

, TIME (SEC)

~

l 1

Fig. 4.1 Acceleration Time History, Perry Reactor Building Foundation (Perry, 1986), (ft/sec 2)

(c) Vertical Component i

i<

0 x 10 4

t.

1

-2.

d 8

-4.

1

~

j l0 3:0 4 .' o s:o el0 TIME (SEC)

Fig. 4.2 Acceleration Time History, Mitchell Lake Road New Brunswick (1982), (ft/sec 2)

(a) 18 0 from North i -,

srl/prrytb1

~

1 o

. X 10 j e.o -

l 6.0 -

4.o -

! 2.o >

1

( y -2.e -

U w - 4 . o- -

^ u k< -e . > -

-e . o- > .

i 4.'o a .'o a .'o

.o a.o X l

l l TIME (SEC) l l

l Fig. 4.2AccelerationTimeHistory,NigchellLakeRoad ,

New Brunswick (1982), (ft/sec ) '

(b) 288 0 from North l

l

a x to

.1 k I o .c alb'[,_ ;'A.%: ; r : M1: : : = :- .; - - - ; - -c

^

g

'l II

=

d _ .i. .

- 8

.e >

.'o 2.'o 4.'o a .'o a .'o X

I TIME (SEC)

G B

Fig. 4.2 Acceleration Time History, Mitchell Lake

. Road, New Brunswick (1982), (ft/sec 2)

(c) Vertical

M

.a.

- a .

e 9 c

.i o

e 1

8 m

W .

s Ed

~~e COO cs O H CfJ A98

. 9OA "s exa

%aO

- ex

%- 8L 45S

. 6 6 6 6 6 6 '9

. 4 M

4 -

? T T k

t e

NOILVB3'1333V 4-16

~ -

.e giiig=-- .

C w

%~

-.__ c s

.o

_ . _u .

H

. . . . . S o 6 6 6 6 6, o 6,

== * ~

?  ?  ?

x NOI1VB3'1333V 4-17

F -

l l

-R R

S S

R t'

e o _ __ o, y --

p ._

y l

&^

de.

m

.)

- 45

• a

- R 832

'4 4J O 8

4 N hSN 8:s

<Qw 9

e

"

• w k 6-g m D a = ,4 e

se- T y

Q.

x

.-4 e

l NOI1VB37333y 4-18

2 I

i x 10 0 2.0--

t U . o- [

! m M

7.

b M

,L -i .e .

e

-2. 0- -

.0 2l0 4.'O 8.0 s.0 10.0 12l0 14.o K

TIME (SEC)

Fig. 4.4 Cumulative Energy (Equ. 4 - 1) with Time, PerryReacg/sec)or 1986), (ft Bg;.lding Foundation (Perry, (a) North-South

1 l

X to s . ~w L

s . o- -

. 5- -

]

g . 0-C

' O cx:

- 14 g .5 .

i 1

! * -1. F -

I

! u o

i:  :  ;

! -1.E  :

s0 s.o 10.0 ta.o 14.0 .

! .o a.0 4.0 X I TIME (REC)

I l

l Fig. 4.4 Cumulative Energy (Equ. 4 - 1)-with Time, PerryReacg/sec)or 1986), (ft Bgilding Foundation (Perry,

. (b) East-West i

I e

, s x

l o

t s

^'-

o a

s ey mr ir Te P

h(

t in

• )

wo i

• C )t E 1a S

( d n

E u 4o N

I T

.F ug f qn a Ei (d

l yi )

gg rBc e e

_ nrs e Eo/l g a

- 4 ectc

- vafi i e(t ta R ,er l y)V ur6 mr8)

• ue9c CP1(

4 4

. g e

o s

x S.

s 0

s

&. r.0

o. .

- 1 s

s.

i F

g :a M $

uiJ~

P

X 10 -

?

4.0--

1

2. 0--

l 1

i

. 0-l O

a:

ta 1 z

. CA -2. 0--

4

-4. o--

.o 2.'o 4.0 a:o alo X

.s.

O TIME (SEC) w i

i

! Fig. 4.5 Cumulative Energy (Equ. 4 - 1) with Time, i hell Lake Road, New Brunswick (1982),

! Mitg/sec)

(ft_ 3 (a) 18 0 from North l

1 srl/prrytbl .

A 10 s.* -

s.* -

4.e -

a.e .

o -

g -a . e -

z

! -4 . 0- -

1

' .b -4. e -

i .

1 to

-a . > - -

.o a:o 4:o e:o e:'

TIME (SEC) i i

i Fig. 4.5 Cumulative Energy (Equ. 4 - 1) With Time, hell Lake Road, New Brunswick (1982),

Mitg/sec (ft ) '

0 (b) 288 from North

. i

s a

X to s

l I

l

. 1--

l l

l .& '

l

( >

o

@ . s- -

m M

. a- -

l o a .'o 4.'o s .'o a.o X

.sti i

i

$ TIME (SEC) 1 Fig. 4.5 Cumulative Energy (Equ. 4 - 1) with Time,

' hell M e Road, New Bmnswick (1982),

Mitg/sec)

(ft 3

(c) Vertical t

srl/prrytbl

s 1

X o

' o.

a 7 _ .

0 5

1

,)

g nc i e Ts g/

tt i f w(

- )

C E

S

) n 1o i

'O. (

t 0

o 7 1 E

M 4M ugn I .

T qi Es (e

yD gn roh eit ntu Eao dS en-

. ' O. vuh 5 iot tFr a o l yN ur mr) uea CP(

6 4

g

- - - i 2

- - F 0

k. -

+6.

.. o-.

+ . &. k.0 1 - - - -

X

>o:amzm b8uno l

tl1 : i !1 1

! l!;;,!.

4 E.

i i'

i I

X to

.e -

I I .e --

.+ -

l

. .o -

1 ,

a 0 i a:

M M

, m  :.. . e- -

L3

< .+.

. e- -

)

. e. -

,, l,

• s:* so:o sslo TIME (SEC) v Fig.

4.6CumulativeEnergy(Equ.4-1)witg/sec)

Perry Foundation Design Motion (ft Ting, (b) East-West .

I

1 I.

l x so a i .& -

i t .& -

I

! .+ -

I

! . 3--

l .C 3

! > . 3- -

o M

y .+ -

ca V

I  ; y ao.o

• s.o so.'o is.o i, 8

.o X i FJ

' 4 TIME (SEC) i i

l i Fig.

4.6 Cumulative Perry Foundation Energy (Equ.

Design4 -Motion

1) witg/sec (ft ) Ting, (c) Vertical .

a 1

=

}

1 i

l 4.3 Response Soectra and Power Spectral Density Functions Figures 4.7, 4.8, and 4.9 show the 2%, 5%, and 7% damped response spectra for the Perry (1986) records, the Mitchell Lake Road, New Brunswick (1982) records and the Perry foundation design motion. These latter components (Perry foundation design motions) do not coincide with the design ground response spectra due to the modeling of the soil by 4

constant soil springs in the CEI reactor building dynamic model. The strong 20 Hz motion in the Perry (1986) records can be readily observed in Fig. 4.7.- The broad high frequency content (above approximately 15 Hz) of the Mitchell '.ske Road, New Brunswick (1982) record is observed from Fig. 4.8. Finally, Fig. 4.9 shows the relatively broad low frequency content of the Perry foundation design motion. These records are not scaled and their peak accelerations are those listed in Table 4.1. "

Powe.' spectral density (PSD) functions were calculated for the nine records--the three components of the Perry (1986) motions, the three

] components of the Mitchell Lake Road, New Brunswick (1982) motions, and the three components of the Perry foundation design motions. The technique used was identical to that of Ref.17. The strong motion duration TD was used which again may be slightly misleading for the

Perry (1986) records. The PSDs were smoothed by means of a three-point j moving average technique which is specified in Ref.19. The results are displayed in two wa~s. y First, the PSDs of the un-scaled and un-

normalized records are shown in Figs. 4.10, 4.12, and 4.14 for the Perry (1986) records, the Mitchell Lake Road, New Brunswick (1982) records,

{ and the Perry foundation design motions. Second, the records were normalized to 1 g. and their PSDs re-plotted versus the US NRC target PSD (Refs. 17-19). For all plots, the units of the PSDs are (in.2/sec.3) to be consistent with the US NRC target.

Figures 4.10 and 4.11 display the distribution of energy as a function of frequency for the Perry (1986) records. Of note is the 4-5 Hz. and srl/prry4.3 4-28 i

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

. . o l

20 Hz. motion especially in the N-S and vertical components. The importance of these frequency ranges was previously demonstrated in Ref. l

20. The E-W component shows broader energy content over the frequency 4

range of interest. The comparison of the normalized PSD with the US NRC I target PSD shows the Perry (1986) records to generally be above the target in the high frequency range.

l Figures 4.12 and 4.13 display the distribution of energy as a function l

of frequency for the Mitchell Lake Road, New Brunswick (1982) records. l They display relatively uniform energy content in the high frequency l range.- Note, these PSDs are calculated and plotted up to 34 Hz as specified in Ref.19. The comparison of the normalized PSD with the US NRC target PSD shows the Mitchell Lake Road, New Brunswick (1982) l records to be above the target in the high frequency range.

Figures 4.14 and 4.15 display the PSDs for the Perry foundation design motion. They have less energy in the high frequency range and, when '

compared to the US NRC target PSD, drop below the target. It is important to note, however, that these records are on the foundation  :

after SSI analysis has been performed so they vary from the free-field design ground motions for which the target was developed. The oscillation or drop below the target PSD in the high frequency range is not uncommon, however, for artificial time histories generated to match RG 1.60 design ground response spectra (Ref.17).

1 l

l l

I srl/prry4.3 4-29 l

l

~.

o x to 2

a) North - South x to b) East - West

" te.c is . 0- - b

$ J H

N '

'ts c .b e 14.0- -

j 3 $2 a-- r l

,3 3* - Iyd y

A { 10.0- 3, b

i . y e.o- . / L g

y hu ,

u /

1. # s.0-- g

.1-- a . o._ Q 2 . 0- -

.c  : .;"...  :

.c .:..... . .::....  : . : : . :'  : . ......

go o I

10 2

0 gg t 10 2 ggi 10 g5 1 10 Frequency (Hz)

Frequency (Hz)-

~

O I x to c) Vertical u 30.C . Legend:

o Perry (1986) 25.o-. 2X dae. ping SX damping _ _ _ _ _ .

$ 20.o-. 7X damping _____. ___ .

' ' ' "' e Notes:

j u A ( All acceleration response spectrum ygo,o ] values in units of f t/(sns) p 5 . 0--

.c  ; .. .: . . .....  : .,

10 10 10

' to l - Frequency (Hz)

Fig. 4. 7: Acceleration Response Spectra. Perry (1986) Reactor prp2t.e Building Foundation Motion.

RSPLT2 V07/05/a5

s 0 a x to e) 18 8

from North x to b) 288* from North 30.c .:

• s

& 25.0-- l g ,

M 3 20.0-- h -

M  %

1 $ ,

. 15.o-- . e- -

A E 10.0-- S b M .#

&  % .1- - ,s -

u(

5. o-- / 9

.c .

.::... .  : : ::::: . ; : ::::' .c  : : :: ...; . ": :'"...  : ..:

10 10 10 10 10 10 10 to Frequency (H7.) Frequency (Hz)

A 8

x to c) Vertical i

Legend: ,

U

1. 0-- Mitchell Lake Rd.

2X damping E , e_-

5% damping _ _ _ _ _ _ .

I 7X damping _ _ _ _ _ _ - _ _ _ _ _ .

e

• S-F Notes:

U .

All acceleration respcnse spectrum

/

% .+-

values in units of f t/(sus)

. e--

.c  : ":':" .  :: ": ": ."

I 0 I 2 ,

16 $0 10 10 Frequency Diz)

Fig. 4. 8: Acceleration Response Spectra. Mitchell Lake Road nbpit.257 RSPLT2 V07/05/85

i -

x 10 0

e) North - S0uth x to ' b) East - West 30.c m so.c et O 25.0--

g 25.0--

" I c 1 l c *. 30. 0--

Nb

• 20. 0-- L e

a Q &)r "

/^ A/

L f e,15. 0- - t' / '

e, 15 . 0-afs~f J e

e u y qhQ g .

W4r"\"%

0 10.0-- fp d < 10 . 0-- $lf jy i

5. 0-- f'

~^ S . 0--

) ^

. ::::: .C  : : : : ::...  : : : . :...: . .. "",

.c ...:.... . ..:.:.. .

10 to

]' to 10 10 to 10 ,

10 Frequency (Hal Frequency Gtz) s d X to # c) Vertical

. 92 Legend:

Perry Design

. s- - g

25 desping 55 demping _ _ _ _

E ,# .

75 desping _. _ .__._. ._ _

l e . 3-- 0 Notes:

r,

/fl o All acceleration response spectre y ,3 f,/

values in units of f t/ (ses) i w

, i- *

.c . .

10 10 10 10 i

Frequency 041) i-i Fig. 4. 9: Acceleration Response Spectre. Perry Design Reactor Building Foundation Motion. pop) l RSPt.T2 V07/05/85 i

s W #

10 M  !!

9  ::

g . .

% j '

M 10g g M,

y'

~

W 10 g g l

,a . .

a g g 10 g g o  !! V f e o

r. 10 g g a  ::

o -

" II a  ::

e w

w  !!

IO i'

! !l 15 . . : . . . .

2 16 10 10 10 Frequency (Hz)

. Notes:

All spectral density values in units of (inuin) / (susus) -

i

Fig. 4.10

Power Spectral Density Function, Perry (1986)

(a) North - South rapped.

ASPLT2 V07/05/85 g

7 4

m so M

g i

N

't3 3. :

N 10 i r M  ::

i N  ::

b . .

t u 10 3 r a .

a g e to y g O

I k

m O a

c. 10 s i:

,o . .-

en

!I 3 .

I . ,

u -

a  !!

II I .

g5 d. . -

0 1 2 si t 10 10 10 Frequency (Hz) ,

Notes:

. All spectral density values in units of (inuin) / (swaws) .

1 1

Fig. 4.10: Power Spectral Density Function. Perry (1986)

A RSPLT2 V07/05/85 (b) Cast - West rapped.l

. ~.

i .

10 ,

in y .E .E g

1 N 3. .

~

'o 30 l l M  :: -

! ,y  :: .

8 A

, 10 g , f u y :i 1 - - -

a

• • g e  !! A o  ::

-  :: f. a

\ e 0 1 C- 10 g g ,

a  ::

u -

. .- l

• 1 1 t . .

s .

I Il u

:

, oi  ::

J

• • 3! !

l

4. . *

[ 15 , , ,

j 15 8

10

.0 10 I

to' Frequency (Hz)

Notes:

I All spectral density values in units of (inuin) / (smsus) -

1 Fig. 4.10: Power Spectral Density Function, Perry (1986) espped, RSPLT2 V07/05/85 (C) Vertical i

s Il m  ::

M $~

H 10 l l N .

y n **t-l .Q  !!

u  ::

y 3 1o _ _ - - - -

N I a

c g.

j N

, 10 g g ' ~ f

E  %

H N I.

N

$ 10 II u  ::

u e

e o-- ll 11 A, .. .

u m ll 3

16 y 15 ' - -

I 0 I 2 16 10 10 10

Frequency (Hz)

Legend: Notes:

i~ All spectral density values in units of Perry (1986) N-S ,

NRC Proposed Target _ _ _ _ . ('inMin) / (susMs) i Fig. 4.11: Power Spectral Density Fp ction. Perry (1986) Normalized rapped.

RSPLT2 vo7/05/85 to 1. g versus NRC Proposed Target (a) North - South

IO l

N  ::

% g.

N M 10 ]- r -

• < :I

. y --~~~~-~

M u

m 10'!!

N F

h c 2"" N l

E 10l l g

.  :: N I N

, 10 hI u  :. :

e O' -

a m 10 l l

?!

4 A,

W t 4  !  : ,

1

! 15 . . . .  : . . .

0 I I I 16 10 10 10 t

Frequency (Hz)

Legend

Notes:

Perry (1986) E-W All spectral density values in units of NRC Proposed Target _____ (inMin) / IsNews) i i

Fig. 4.11: Power Spectral Density Function. Perry (1986) Normalized RSPLT2 V07/05/85 to 1. g versus NRC Proposed Target (b) East - West rapped.:

)

m w lr N

% * *g-U I!

N ..

a

• • 'l !

>. g- u u 10 g, 4 ,.

E :5 "J ,

! S 8' ( [ N f

% g

, N a

10 n-- s c.

o  !!

u  ::

o g-2 ll y ..

1 i

u  ! l

= 5i I!

I l

C iS . .-

.I 16 10 10 10 Frequency (Hz)

Legend: Notes:

Perry (1986) All spectral density values in units of Vertical . (inmin) / (susMs)

NRC Proposed Target - _ _ _ .

! Fig. 4.11: Power Spectral Consity Function. Perry (1986) Normal'ized RSPLT2 V07/05/85 to 1. g versus NRC Proposed Target (c) Vertical rappse

1 4

10 y ,

m  ::

et  ::

H N S. .

'O 10 g g

• M ii 4s  :: .

E

! . 10 g g '

W A ii e g

F c 10 e ll .

O  ::

e 4

O

c. 10 g g i a  :  :

, u -

~

~

e ~ ~

at

' " II l

i

.?.?

s g g u  :  :

16 g g

-4 0.0 .: .-

10 . . . . . . . .

I 10 0

10 I

10

. 16 Frequency (Hz)

Notes:

All spectral density values in units of (inmin) / femens) -

i l

i 4.12: Power Spectral Density Function, Mitchell Lake RJ Fig.

06) 18 from North esppsc RSPLT2 V07/05/85

88 i [

1 m f:

l<

n p

\ 10 m I l n  ::

• 10 e..

=

> . I  !

u o  :.. :

e o

i ..

e 0 i c to b.5 a -.-

o --

e - -

i n

II I

b

8 1 a  : [.

.i O  ::

~~

-3 f.

to y  ;

s d

s5 .

0 I 10" l 10 10 ti l

4 I Frequency (Hz) l Notes:

All spectra 1 density values in units of finninl/fs*s*s) l s Fig. 4.12: Power Spectral Density Function, Mitchell Lake Rd espped.

l l FISPt.T2 V07/05/85 (b) 288 from North i

l 4

d 10 l  !!

1 i M 3-~

P 10 N !I m  :: A M

a

I ,

i Q^ 10 g g j e p u --

a g C

to i l -

O i .i i

m O

c. 10 g g ,

! u  :: -

o, . .

!\

l i .! ! .

a  ::

p . .

, -3 10 g g ~

s .i i, .

j 4 .:.  : : :...

, 10 .  : . ::: . . .

1 0 I 8 10 10 10 10 Frequency (Hz) 1 4

! Notes:

I All spectral density values in units of i .

[

(inain) / (sasas)

.i

. Fig. 4.12: Power Spectral Density Function. Mitchell Lake Rd rapped.l i RSPLT2 V07/05/85 (c) Vertical l

4

~.

I ll m  ::

N 5 H 10 i

N ll

'o M g -

M 10 N 11 A-

,,r-! _ _ _ _ _ . - - - ~

N a .! .  %

N c 2-N

[2 w 10 l l N N

N i

E 10 I::

a u

!I e e-0 10 11 e

b u  !!

1

-Y-IO i .! !

,i 16

1. ~  :  : .:  :

I go t 0 1 I 10 sI 10 Frequency -(Hz) .

1 Legend: Notes:.

Mitchell Lake Rd All spectral density values in units of l 18- from North (inMin) / (SMSNS) i NRC Proposed Terget _ _ _ - .

l 1

Fig. 4.13: Power Spectral Density Function. Mitchell Lake Rd rappsd.

j RSPLT2 Vo7/o5/85 Normalized to i.g vs. NRC Proposed Terget (e) North

'1 i

1

II

. m  ::

n 5- -

H 1 N 10 l

't3 :li M g-M* IO l l y gg i e :i N s

c g- -

• 10 N O ll

N

- N I-s f 10 g y N

+8  ::

u - .

e o--

i n co 10 l l

i l b

! ll l

i 8

l 15! "!

1

-c-

! 10 .  :. . . .  : .: .  : . .:

16 0 I 2 10 10 10 Frequency (Hz) l Legend: Notes:

! *Mitchell Lake Rd All spectral density values in units of 288 from North i

(inmin) / (sNams) i NRC Proposed Target - _ _ _ .

1 l Fig. 4.13: Power Spectral Density Function. Mitchell Lake Rd j RSPLT2 V07/05/85 Normalized to 1.g vs. NRC Proposed Terget (b) West rapped.:

1 .

l

n ll m  ::

M T' s to j g:

N y .

I M 4"~

4u .! .!

,,r 11

_______---s~ f N

E ji J

$ 18 1 1  %

s

-  : N I N -

E 10 g g u .

E **,! !

11 A  ::

i ..

A I 16 )

[

15 ' . ...: "

10 10 10 10 .

Frequency (Hz)

Legend: Notes:

Nitchell Lake Rd All spectral density values in units of Vertical (inWin) / (smsus)

NRC Proposed Target ----.

Fig. 4.13: Power Spectral Density Function. Mitchell Lake Rd RSPLT2 V07/05/85 Normalized to 1.g versus NRC Proposed Target (c) Vertical rapped.

I

,\1i\ i \\;i\ \t\;\\ i

\\\

9 '

d e

p p

s e

a e

s n

g i

s e

D f Y o r r

s e t P

., i n n

u o

i n - t i

) c z s n H

( e u u F y l y c a t n V i e y s u

q t n i . eh e s Dt r n u F e) lo d s aS a r o t -

o l s c t au eh

. ra pt

t( sro c/

e) pn rN s si e e w w) oa t l n p(

o li N A(

I d

i d

i 9 4 ui i : - -

• g:. - a: -

y:; . {:. . 1:. . t: 5 !i -

t. 1 8 o s 0

. . 1:. . I: 3 5 !E 4

. F

$ s o

s 0 si ,o 1

1 s

>a 2c=O ,Euu.a*

s e

/

5 0

/

7 c

v 2

a.u a1* T R

s n

\ s' \l >

e

~.

1 4

! 10 , ,

I

=

:

et ..

H

'3 i 4Pt

~~

M I

2 a 10 g g

. >=  :

u 8 w  ::

i - a g"~ .

o "

s ..

e 0

c. 10 y g a  ::

u -

,o ..

• l t:

i ..

e 1I .

a  ::

• .~.~

8 18 p 15 4.. .

I 0 I I

16 10 10 10 .

Frequency (Hz)

Notes:

All spectral density values in units of (innin) / (susus) i l

Fig. 4.14: Power Spectral Density Function. Perry Design RSPt.T2 V07/05/85 (b) East - West espped.

d to y ,

m ..

n ..

( 10 g f

'd  :

n .- .

g .. I M 10 L o  ::

- ~

a g 4

c 10 1

e 1 1 i

o  :. :.

e o L 10 g g a  :  :

o  ;;

l E --

i 1 1 a II a  ::

! a .-.-

q . .

-3 10 g g j .  ::

-4 so .-

l ggi 0 g 2 10 10 to 1

Frequency (Hz)

Notes:

i All spectral density values in units of (inwin) / (susus) l M

Fig. 4.14: Power Spectral Density Function. Perry Design j RSPt.T2 V07/05/85 (c) Vertical rapped.p i

l 1

w II m  ::

n g-P 10 N Il e  ::

n t- .

n 10 N !I a  ::

i -

r -

-s's m

! - ,p !

1 m . .!  %

l c g- -

• N I O 10 l l N I

y :E -

N J e 1-  %

(- 10 g ,

1 a . .

o

:

I e a-I 2 So ! I

l .l

. a i  ::

6 m 11 I'

1 15 i 11

]' to( -,

l 10 10 10 10 l

Frequency (Hz)

Legend: Notes:.

Perry Foundation All spectral density values in units of Design N - S (inuin) / (suswe)

NRC Proposed Target _ _ _ _ .

1 I

i

-l l

l Fig. 4.15: Power Spectral Density Function, Perry Design Normalized i RSPLT2 V07/05/85 to 1. g versus NRC Proposed Target (a) North - South rapped.

+w

~.

ll

! m  ::

n T-w to g r 7

M i 4,! *! .! .

,,r-l _ ___ _ _ _

% ~ t

- I N l

e  :: s P c y-. s 10 N S 11 s e

fi -

N a t- ,

c.

o to I y:

u . .

e 8*0"-

s  !!

11 A, . .

! A e ll

{

_r - ,

i . 10 l l 1 fi (

1

! 10( - . .

I 8 E 2 16 10 10 10 Fr%quency (Hz)

Legend: Notes:-

i

. Perry Foundation All spectral density values in units of

Design E - W (inWin) / (susus)

NRC Proposed Target ____.

i lJ .

l l

Fig. 4.15: Power Spectral Density Function. Perry Design Normalized j RSPLT2 V07/05/85 to 1. g versus NRC Proposed Target (b) East - West r spoed.:

i b$

M I

w 10 \: :\

, N b > j l

y 3-- ___ ~~~~

" -* gg ! ! *

! g to==g g

's N tn

)y SE 1

E u 10]  !:

i u - -

e .

• 0--

E  !!

-~

l .

un gr O  !!

AI ] [

16 . .  : .: . . . .  : : :

I .0 I E 16 '10 10 $0 i Frequency (Hz) l Legend: Notes:.

' Perry Foundation All spectral density values in units of Design vartical (inmin) / (smsus)

NRC Proposed Target ____.

1 Fig. 4.15: Power Spictral Density Function. Perry Design Normalized RSPLT2 V07/05/85 to 1. g Versus NRC Proposed Target (c) Vertical espped. pit

l 4.4 Nonlinear Resnonse of Sinole Dearee-of-Freedom Systems -

t Past history as documented by Kennedy et al. (Ref. 4) has shown that linear elastic analysis has not led to good predictions of structure behavior when failure is the principal' interest. In particular, the damage potential of earthquakes is not well-described by low-damped elastic response spectra. Kennedy et al. Indicated that wel?-designed structures can easily survive grour.d motions 2.5 times the motion that would cause yield in structural members. Inelastic response of shear

[ walls begins as soon.as extensive concrete cracking occurs. For heavily i

reinforced shear walls (steel percentages of about one percent) typical l of those found in nuclear power plants, inelastic behavior occurs before' the code specified minimum capacity is reached. Kennedy et al. assessed the damage potential of ground motions considering nonlinear behavior of simple single degree-of-freedom systems. Their investigation sought scale factors by which earthquake records must be scaled to induce i

specified levels of nonlinear deformation. Two levels of nonlinear deformation were considered, as defined by ductility ratios of 1.85 and

4.27. A ductility level of about 1.85 represents a reasonable lower estimate of the inelastic deformations which would occur in a shear wall i

designed for static lateral loads to the ACI - 349 code ultimate

~

capacity. Current elastic design analysis utheds when carried to code .-

ultimate capacities are judged to lead to roughly this level of inelastic deformation.

j Kennedy et al. analyzed single degree-of-freedom models representative i

of the fundamental frequency characteristics of stiff structures with elastica 11y calculated frequencies varying from 2.14 Hz to 8.54 Hz. The inelastic behavior of these models was described by a shear wall model exhibiting stiffness degradation after yield and pinching of the hysteresis loop during loading direction reversal. The stiffness of the shear wall model beyond yield was taken to be 10% of the elastic stiffness. A modified version of the computer program DRAIN 20 as 3

described in Ref. 13 was used to perform the analyses. The eleven

/

l 4-51 i

• l a'

• 1 recorded motions and the artificial acceleration time history were considered in their study. A result of their analyses was scale factors which must be applied to each record and for each structure model to achieve deformations corresponding to ductility ratios of 1.85 and 4.27.

The resulting scale factors may be used to scale the records up to achieve the desired result or their inverse may be used to scale down the record and estimate an effective peak acceleration. Note the scale factors are dependent on the specific records, the desired ductility level, and the structura, model elastic frequencies.

4 A similar study was performed here to assess the potential effects of the Perry (1986) recorded motions on the Perry structures. The study followed the procedure used by Kennedy et al. (Ref.13). Two single degree-of-freedom systems were considered. Their elastic frequencies were 5.4 Hz. and 8.9 Hz. approximating the fundamental frequencies of i

the Perry drywell and the Perry auxiliary building, respectively. These modes were considered representative of horizontal behavior. In the a

present study six records were considered--the two horizontal components of the Perry (1986) records, the two horizontal components of the Mitchell Lake Road, New Brunswick (1982) records, and the two horizontal e

components of the Perry foundation design motion. The nonlinear shear wall stiffness element described above was used. The yield force value was developed based on the following assumptions. The structure was ,

designed to a force level corresponding to the spectral acceleration of i

the design ground response spectra (Regulatory Guide 1.60) at the j

fundamental frequency of the structure and for 7% damping. The design

{

force level leads to a deformation corresponding to 1.85 times the yield

[ displacement.

The stiffness of the shear wall model beyond yield is 10%

! of the elastic stiffness. Nonlinear time history analyses were then performed as described in Ref.13. The analyses *were performed for the portions of the time histories defined by the strong motion duration T D. A series of nonlinear analyses was then performed to calculate the scale factors necessary to scale the records and achieve nonlinear j

deformation corresponding to 1.85 times the yield displacement, i.e., to achieve the deformations expected when the design level forces are

{ 4-52

, s. .

deformation corresponding to 1.85 times the yield displacement, i.e., to

{

achieve the deformations expected when the design level forces are reached for the records considered. Table 4.4 contains the results.

For the 5.4 Hz structure, the Perry (1986) N-S and E-W components would need to be scaled by 5.3 and 5.5 respectively to induce desian level deformations in the structure. Ground motion acceleration time histories identical to the Perry (1986) recorded motions must be scaled to peak accelerations of 0.95g for the N-S component and 0.579 for the E-W component to induce design level deformations in the 5.4 Hz simple structural representation. Alternatively, an effective peak acceleration of these records and for the 5.4 Hz structural model would be 0.034g and 0.018g for the N-S and E-W components, respectively.

Table 4.4 presents results for the twelve cases considered. Note, the definition of the scale factor calculated here is slightly different than the scale factor reported in Ref.13. Here, both elastic and inelastic response contribute to the scale factor. The elastic portion scales the 7% damped spectral acceleration of the record of interest at the frequency of interest to the design response spectra. The inelastic portion is the, scale factor necessary to achieve the specified level of nonlinear deformation. The inelastic portion varied from 1.57 to 2.16 for the recorded motions (Perry,1986 and Mitchell Lake Road, New Brunswick,1982). The scale factors for the Perry foundation design motion are near one.

4-53

I 1

Table 4.4 Scale Factors to Achieve Ductility Ratios of 1.85 and Corresponding Effctive Peak Accelerations for Selected Acceleration Time Histories i

1 l

l Earthquake Model Structure Frequency *

! Record 5.4 Hz 8.9 Hz

, Scale Eff Scale Elf ,

]. Factor Accel. (g) Factor Accel. (g)

I Perry, 1986 5.3 .034 6.7 .027

) (N-S)

) Perry, 1986 5.5 .018 4.3 .024

! , (E-W) i i 1 T Mitchell Lake Rd 4.9 .031 3.3 .046 i New Brunswick,1982, 0

(18 from North) j i

Mitchell Lake Rd 3.6 .067 2.8 .086 New Brunswic):, 1982 0

(288 from Horth)

Perry Foundation 1.2 .15 1.1 .16 l Design Motion

(N-S) s .

l 1  !

j Perry Foundation 1.2 .14 1.1 .15 i Design Motion j (E-W) 4

a e

5.

SUMMARY

AND CONCLUSIONS A summary of the results of the independent studies performed to evaluate the effects of the January 31, 1986 Ohio earthquake on the Perry Nuclear Power Plant structures is as follows:

a The total. energy of the Perry (1986) recorded motions was calculated for each of the three components. They were compared with the energy in fourteen other recorded motions (eleven from a previous study and three for the Mitchell ' Lake Road, New Brunswick,1982 earthquake), and four artificial time histories (one from a previous study and the three components of Perry foundation design, motion). The energy in the Perry (1986) records was less than that in any of the records considered--much less for the longer duration motions. Along with the total energy, the strong motion durations, power, and RMS accelerations of the records were c: tim;ted. Relatively short duration, low perer, .d low RMS accelerations were calculated, a

Response spectra and PSD functions of the Perry (1986) three components of motion, the Mitchell Lake Road, New Brunswick (1982) three components of motion, and the Perry foundation design motions were calculated. The PSDs showed the concentration of energy in the high frequency range for the two sets'of recorded motions.

Comparing the normalized PSDs to the US NRC target PSD shows their values to be above the target in the high

~

frequency range.

s Nonlinear analyses were performed on single degree-of-freedom models representing the fundamental horizontal srlprry 5. -

5-1

. c frequency of the drywell (5.4 Hz) and the auxiliary -

buil. ding (8.9 Hz). Scale factors were calculated which when applied to the input motions would achieve a nonlinear deformation in the simple model of 1.85 times the yield displacement. This level of nonlinearity is expected when typical concrete shear walls are loaded to their desian level. The scale factors calculated here included an elastic and inelastic portion. The scale factors for the Perry (1986) records were: (5.3,b.5) for the (N-S, E-W) components and the 5.4 Hz model; and (6.7, 4.3) for the (N-S, E-W) components and '8.9 Hz model. A measure of the effective peak acceleration of these records is the instrumental peak accelerations divided by the scale factors.

1 d

a The SSI and structure response aspect of the .

investigation confirmed that SSI was not an important phenomenon for the rock-founded Perry structures.

Neither kinematic nor inertial interaction appears to have occurred to a significant extent. The effect of non-vertically incident waves on the response of the containment vessel was assessed. Based on in-structure response spectra in the containment vessel, vertically incident waves is the most likely wave propagation mechanism, a Comparing predicted and measured response in the containment vessel at elevation 688', the frequency characteristics of the N-S, E-W, and vertical directions correspond well with one exception, i.e., the low frequency E-W component. Response predictions are within 30-35% with variations being primarily an under-prediction for E-W and vertical. It is well-recognized that peak spectral amplifications are uncertain, hence,  ;

the match is adequate.

srlprry 5. 5-2

a O Based on thlse studies, a revicw of CEI's evaluations and other '

technical information, the January 31, 1986 earthquake is judged to have had an insignificant effect on the Perry Nuclear Power Plant structures.

Further, it is judged that the Perry seismic analysis models adequately predict the behavior of the reactor building when subjected to this event. Although, it is recognized that a portion of the high frequency motion recorded on the containment vessel may have been due to secondary effects, such as polar crane vibration or impact, and this remains an open question. Finally, the plant design of the structures is judged to be acceptable and unaffected by the event.

The effect of the January 31, 1986 on equipment was assessed by the US NRC in a separate effort. '

l l

srlprry 5. 5-3

> o .

. 6. REFERENCES

1. The Cleveland Electric Illuminating Co., " Seismic Event Evaluation Report, Perry Nuclear Power Plant, Docket Nos. 50-440; 50-441," February 1986.

2. Gilbert / Commonwealth, " SAP IV Input and Output Listing-Perry Reactor Building Updated Seismic Analysis,"

Received February 17, 1986.

3. The Cleveland Electric Illuminating Co., " Seismic Event Evaluation, Technical Presentation, February 11, 1986,"

presented at the Perry Nuclear Power Plant, February 11, 1986.

4. Letter M. Edelman to H. Dcnton, " Perry Nuclear Power Plant Docket Nos. 50-440; 50-441 Seismic Event Evaluation Report" Sunnlemental Information. February 28, 1986. Attachments 1-5. Attachment 3 Equipment Seismic Qualification Evaluation, PY-CEI/NRR-0438L.

5. Letter M. Edelman to H. Denton, " Perry Nuclear Power Plant Docket Nos. 50-440; 50-441 Seismic Event Evaluation Report" Sunnlemental Information. March 3, 1986, PY-CEI/NRR-0440L. .

6. Transmittal, C. Chen to J. J. Johnson, "IDI Package and Original Response Spectr,a," March 18, 1986.

7. Letter M. Edelman to H. Denton, " Perry Nuclear Power Plant Docket Nos. 50-440; 50-441 Seismic Event Evaluation Report Suonlemental Information, March 11, 1986, PY-CEI/NRR-0442L.

8. Letter J. J. Johnson to R. Hermann, " Questions on CEI Submittals and Previous CEI/NRC Meetings - Revision 1,"

April 28,1986.

9. Memorandum, C. Chen to J.'J. Johnson, Responses to Ref.

8, May 22, 1986.

10. Gilbert / Commonwealth, Inc., "The Cleveland Electric Illuminating Company Perry Power Plant Confirmatory Program of the January 31, 1986 Ohio Earthquake Effect Docket Nos. 50-440; 50-441," Draft, June 9, 1986.

11. Gilbert / Commonwealth, Inc., "The' Cleveland Electric Illuminating Company Perry Power Plant Confirmatory Program of the January 31, 1986 Ohio Earthquake Effect l

l 6-1 l l

m _ _ _ . . _ _ . _ _ ,_

, s Decket Nos. 50-440; 50-441," June 16, 1986, G/C Report No. 2632.

12. Wong, H. L., and Luco, J. E., " Soil-Structure Interaction: A Linear Continuum Mechanics Approach (CLASSI)," Dept. of Civil Engineering Univ. of So.

Calif., Los Angeles, CA, CE 79-03,1980.

13. Kennedy, R. P., Short, S. A., Merz, K. L., Tokarz, F.

J., Idriss, I. M., Power, M. S. and Sadigh, K.,

" Engineering Characterization of Ground Motion, Task I:

Effects of Characteristics of Free-Field Motion on Structural Response," NUREG/CR-3805, Vol. 1, Prepared for U. S. Nuclear Regulatory Commission,1984.

14. Arias, A., "A Measure of Earthquake Intensity," Seismic Desian for Nuclear Power Plants, MIT Press, Cambridge, Mass., 1970.

15. Housner, G. W., " Measures of Severity of Earthquake Ground Shaking," Proceedinos of the U.S. National Conference on Earthauake Enaineerina, EERI, Ann Arbor, Mich., 1975, pp. 25-33. -

16. Trifunac, M. D. and Brady, A. G., "A Study of the Duration of Strong Earthquake Ground Motion," Bulletin of the Seismoloaical Society of America, Vol. 65, 1975, pp. 581-626.

17. Coats, D. W. Jr., and Lappa, D. A., "USI A-40 Value/ Impact Assessment," Lawrence Livermore National Laboratory, Livermore, CA, Prepared for the US NRC, NUREG/CR-3480,UCRL-53489,1983.

18. Shinozuka, M., " Recommended Position Statement Concerning NRC R.G. 1.60 Requirements," Unpublished Paper, May,1983.

19. US NRC, " Standard Review Plan Sec. 3.7.1 Seismic Design Parameters Working Draft Rev. 2, 1986.

20. Weaver, H. J., and Burdick, R. B., " Spectral Analysis of Perry Nuclear Power Plant Velocity--Time Histories,"

Lawrence Livermore National Laboratory, Livermore, CA, August 1986.

21. Gilbert / Commonwealth Inc., "The Cleveland Electric Illuminating Company Perry Power Plant Confirmatory Program of the January 31, 1986 Ohio Earthquake Effect," Progress Report, April 30, 1986.

6-2 f\