ML13330A080
| ML13330A080 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 08/08/1980 |
| From: | Koss P Southern California Edison Co |
| To: | |
| Shared Package | |
| ML13330A079 | List: |
| References | |
| TASK-03-06, TASK-3-6, TASK-RR NUDOCS 8008150199 | |
| Download: ML13330A080 (17) | |
Text
A COMPARATIVE EVALUATION of the SAN ONOFRE SITE SPECIFIC SPECTRA with the 2/3G HOUSNER SEISMIC REEVALUATION SPECTRA Paul Koss August 8, 1980 8-0080150
TABLE OF CONTENTS
1.0 INTRODUCTION
2.0 BASES FOR COMPARISON 3.0 VERTICAL SPECTRA 4.0 HORIZONTAL SPECTRA
5.0 CONCLUSION
6.0 REFERENCES
1.0 INTRODUCTION
The purpose of this duscussion is to compare the site specific spectra (including the Imperial Valley refinements) with the seismic reevaluation program criteria, which is based upon a 2/3g Housner spectrum in the horizontal direction and a 0.44g Housner Spectrum in the vertical direction.
The seismic reevaluation program may be considered in two parts. These are:
(1) The seismic backfit project (SBP) which was completed in 1977 and is described in reference 1 and (2) the balance of plant seismic re evaluation (BOPSR) which is currently being performed. During the SBP the reactor building, containment sphere and NSSS were reevaluated. The NSSS supports were modified to provide a minimum margin of 10% against allowable stresses in the supports, equipment, and piping. The scope of the BOPSR includes the remaining safety related structures, equipment and piping.
The principal features of the overall seismic reevaluation program include the use of state-of-the-art analysis techniques including: Three dimensional system models, nonlinear analysis, simultaneous input of three orthogonal ground motion components, system-subsystem interaction, and modeling of structures and components based upon as-built information. Allowable stresses are established based upon (1) Applicable codes at the time of the original design (2) quality control procedures in effect during con struction and (3) current code requirements.
The San Onofre site specific spectra, including the Imperial Valley re finements, are shown in figures 1 and 2.
Also shown in figures 1 and 2 are the 2/3g Housner and 0.44g Housner horizontal and vertical response spectra.
The site specific spectra represent a mean plusf1 plot of ahalytical results.
For the purposes of subsequent discussion the mean plus107plot of the site specific spectra, including Imperial Valley refinements, will be referred to as the site specific spectra.
2.0 BASES FOR COMPARISON The comparison of the seismic reevaluation design criteria with the site specific spectra must be made separately for structures and equipment. This is due to the fact that the 2/3g Housner spectrum is input directly into the analysis of structures while the analysis of equipment requires (1) The development of a synthetic time history which conservatively envelopes the 2/3g Housner spectra, (2) computation of time history responses and response spectra at equipment support points within the structures and (3) the develop ment of a smoothed response spectra (which conservatively envelopes the com puted time history response) for input to equipment analysis. Furthermore, if the equipment is to be qualified by testing in lieu of analysis the smoothed floor spectra are conservatively enveloped by the test input.
In view of the method of analysis of equipment, the most appropriate method for assessing the relationship between the site specific spectra and the seismic reevaluation criteria is to compare the synthetic ground motion time history spectra with the site specific spectra at the fundamental periods of the structures. This will give a rough approximation of the relationship between the maximum floor acceleration due to the synthetic time history and maximum floor acceleration corresponding to the site specific spectra.
3-
3.0 VERTICAL SPECTRA The vertical site specific response spectra are enveloped by the vertical 0.44g Housner design response spectra (see figure 2) at damping values greater than 2%. Since the damping in structures exceeds 2%, the 0.44g vertical Housner spectra is a more conservative input for the design of structures than the site specific spectra.
The input to the design of equipment is the structural response to the synthetic time history. The vertical response of each structure will generally be based upon the period of the fundamental vertical mode of the structure and its damping. Higher frequency modes generally don't contribute substantially due to the dominance of the soil structure inter action modes. For any range of fundamental mode periods the synthetic time history will probably produce a higher vertical response in the structures than would be produced by the site specific spectra because at structural damping (greater than 2%) the vertical synthetic time history spectra also envelopes the site specific spectra (see figure 5).
Based upon the above considerations it can be concluded that the vertical
.44g Housner design spectrum is a more conservative design input than the site specific vertical spectrum.
4.0 HORIZONTAL RESPONSE SPECTRA A comparison of. the site specific spectra to the 2/3g Housner horizontal design spectra is shown in figure 1. At periods of less than 0.3 seconds the 2/3g Housner design spectrum exceeds the site specific spectra by as much as 60%.
At periods of 0.3 to 0.6 seconds the site specific spectra exceeds the 2/3g design spectrum by a maximum of about 18%.
During the SBP, the fundamental periods of the reactor building and containment sphere were computed. These computations indicated that for a DBE these structures had fundamental periods between 0.2 and 0.3 seconds (see reference 1) therefore, the 2/3g Housner spectrum was a more conservative design input for the reevaluation of the containment and reactor building than the site specific spectrum. A similar conclusion may be reached regarding the NSSS equipment by comparing the synthetic time history spectra at the fundamental periods of the supporting structure (0.2 to 0.3) to the site specific spectra (see figures 3 and 4).
For example, at a period of 0.25 seconds (roughly the rocking soil structure inter action mode) the 7% synthetic time history spectrum exceeds the 7% site specific spectrum by about 30%. From this it can be concluded that the 2/3g Housner synthetic time history will produce a floor acceleration whose amplitude would be about 30% greater than what would result if the site specific spectra were the basis for computing floor response spectra.
It can be concluded from the preceeding discussion that if the site specific horizontal design spectrum were to be substituted for the 2/3g Housner design spectrum, the calculated margins in the spherereactor.building, and NSSS that were reported in reference 1 would increase significantly. Therefore, a
comparison of the 2/3g horizontal Housner design spectrum with site specific spectrum indicates that the conclusions presented in reference 1 regarding the seismic adequacy of the containment, reactor building and NSSS are conservative.
The BOPSR project includes the analysis of structures and equipment that were not part of the SBP. Since this work is in its formative stage, the periods of the fundamental modes of these structures are not available to be used in comparing response spectra. It can be observed, however, that the 2/3g Housner spectra are more conservative than the site specific spectra in a range of periods from 0.0 to 0.3 seconds and greater than 0.6 seconds.
In the range of 0.3 to 0.6 seconds the site specific spectra exceed the 2/3g Housner by amounts varying from 0.0 to a maximum of 18% at a period of 0.4 seconds. This difference is not very significant since a structure with a period of 0.4 seconds which is near the elastic limit for a 2/3g Housner design spectrum would only require a ductility of about 1.2 to have sufficient reserve capacity to withstand an input 1
equivalent to the site specific spectra at this period.
This, of course, assumes the structure is at or near the elastic limit. Actually, however, most structures have some margin against allowable stresses (as was observed during SBP) and the allowable stresses generally have some margin against the elastic limit. In addition to ductility, damping and actual insitu material properties will tend to offset minor differences in spectral amplitude.
A review of figures 3 and 4 demonstrates that the synthetic time history spectra conservatively envelope the site specific spectra at most periods and that only a few points fall below the site specific spectra curves. On this basis it is concluded that when floor 1 The method for determining the effect of ductility ratio on reserve capacity is described in section 7.4 of reference 2.
response spectra are computed and broadened in accordance with the methodology described in BC-TOP-4A, a conservative basis will be established for the analysis of equipment and piping.
Based upon these and other considerations it is concluded that since the 2/3g Housner is significantly more conservative (up to 60%) than site specific spectra over a broad range of frequencies and is slightly exceeded in only a narrow frequency band, it represents a conservative design basis for the BOPSR.
5.0 CONCLUSION
The site specific spectra have been reviewed and compared to the 2/3g Housner design spectra, which are currently being used in the seismic reevaluation program. The conclusions of this report are based upon a comparison of the spectra curves and an analysis of their relative impact upon the design process. No attempt has been made to reduce the site specific spectram to an effective acceleration by taking into account wavelength and inertial effects.
By comparing the site specific spectra and the 2/3g Housner spectra in the period range of the containment and reactor building it was concluded that the 2/3g Housner is a more conservative seismic input. A similar conclusion was reached by comparing the synthetic time history-spectra with the site specific spectra. As a result, the only significance of the site specific spectra on the work completed in 1977 under the seismic backfit project (SBP) is that they demonstrate the conservatism of the seismic input used in the analysis.
The Balance of Plant Seismic Reevaluation (BOPSR) program will utilize the 2/3g Housner spectra,. which are generally more conservative than the site specific spectra.
The small exceedence of the 2/3g horizontal Housner spectra in the period range of 0.3 to 0.6 seconds would very easily be offset by a small amount of ductility, increased insitu material strength, or slight increases in damping. In addition, the synthetic time history spectra envelope the site specific spectra sufficiently to conclude that when the computed floor spectra are enveloped and broadened a conservative basis for the seismic re evaluation of equipment and piping will be established.
Based upon the above considerations it is concluded that the 2/3g Housner spectra provide a conservative seismic input to the seismic reevaluation program.
6.0 REFERENCES
(1)
"Seismic Reevaluation and Modifications", San Onofre Nuclear Generating Station, Unit 1, April 29, 1977, USNRC Docket 50-206.
(2) Newmark, N. M. and Hall, W. J. "Development of Criteria for Seismic Review of Selected Nuclear Power Plants", (NUREG/CR-0098),
May 1978.
(3) Hadjian, et al, Seismic Analyses of Structures and Equipment for Nuclear Power Plants, BC-TOP-4A, Revision 3, November 1974.
isi Al of 7 9 1
! I I I Ii7 I0 D
LL~~~~~~
si6/6 olol&ew i
A 51eS0EIF Sfr I;R4 1.2ol~
s-Ec
.m S1____________
74
- CIA C
R OAS4 P
C7/aW 6~~~~~~~~~~
0I7 SECrk45675 fE6OvvA7
~
is 4
7 7
T7~3 1.1 I
I L___________
CA 10 a
c I--
c t
VR
T -1*
1.1 1.1*
1.2 1.0
.e%
SPECTRA SONGS I
sicE'FIEL HOIZOTAL LL!
1*M 4,
150A
-~c HI IZI
~fli I
411 LD.
1 00 7 C 4
'YA P E 01 D
(, SE
)-1 7
,,e
-f4,
~~~~~~~~~Sp C ______
______ i~~
r oq TRACE__
A 20SE.
T.
01 EC ST FPCFCSECRMFO/FN4 E
F/6/.
- 3.
COPI/3/VOFTH
/0r/4
__f___GE_
FYA/ZHE T/C_7 HiISTR PC/N
-13
s'n 1.8 1.7T K.>2 P
1 0 D
S SEC SPECTRA SDNGS 1 FREE. FIELD HORIZONTAL I-TACE B
20 SEC.
AT 0.01 SEC.
L F/15.RE 4 COMPARIS0ON Or THRE Hllov
-5YA/./7/E 7,1c r1/ME., HI~s7rory sRr7F4Mv S/TEF 5'PtC/rC SPECrcq.
- 1.
1 2._______
__I__
- 1.
L..: L I
T j
h Ii I ~ I-.......-..
I i
- 1E R___
_I CA___e LL S
Y_______
A-1 7_1_C___________
rF slc______
- 5.
C0M__________OA-rSVE_71_.IZ 0-44, S
_____ _____ _____ _____ _____ ______ _____ ____R__yn o_
r
_15_
-