ML20118D317
| ML20118D317 | |
| Person / Time | |
|---|---|
| Site: | Bodega Bay |
| Issue date: | 06/26/1963 |
| From: | Newmark N AFFILIATION NOT ASSIGNED |
| To: | Bryan R US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML093631134 | List:
|
| References | |
| NUDOCS 9210120104 | |
| Download: ML20118D317 (15) | |
Text
a
.f-
.~..,
~ s
's j f6
- 2(g~
~
NATHAN M.
NEWMARK Itg.,
G,.s f c y M '-
17o5 S.
PLEASANT S T... U d B 'A N A ? l L L I N O I SO 26' June 1963" Mr. Robert h. Bryan, Chief Research and Power Reactor Safety Branch Division of Licensing and Regulation V. S. Atomic Energy Commission Washington 25, D.C.
Re: Transmittal of Draf t of Comments on Bocega Bay Reactor
Dear Mr. Bryan:
Enclosed are two copies of the draft of my comments which I hope reach you in advance of your departure for our conference in Chicago on 2 July.
I am sending one copy directly to Mr. Williamson.
I shall arrive in Chicago via Ozark Flight 736 at 5:40 p.m.
on Monday, 1 July, and I shall come directly to the Holiday Inn.
I expect that over dinner and af terward we can get our discussions taken care of in advance of the meeting on Tuesday.
I have to get a return flight at 5:00 p.m. on Tuesday cf ternoon.
With be=t regards.
Sincerely, k
LIW\\o.v N. M. Newmark NMN:bJw Enclosure QN _, b- ]h cc:
Mr. R. A. Vi11iamst t_
h)RC'
,5 odDECr A Mq jocKET So - 20 p
b 9;3ac Ps r
53M oN g I
rutere A" e s, Aa(s 7tAE +
A3 LTRs* hEg
$Tgte 9IRO 8% U i
9210120104 920520 F
L643 PDR ORG NitCH s.v
~ i',h 5] - *
l fg-gg
~
NATHAN M. NEWMARK I&R p 3,
dM 17 o s s.
PLEASANT ST..
U R B'A N iE/ l L L I N O I SM 26 June 1963 Mr. Robert H. Bryan, Chief Research and Power Reactor Safety Branch Division of 1.icensing and Regulation U. S. Atomic Energy Commission Washington 25, D.C.
Re: Transmittal of Draf t of Comments on Bodega Bay Reactor
Dear Mr. Bryan:
Enclosed are two copies of the draf t of my comments which I hope reach you in advance of your departure for our conference in Chicago on 2 July.
I an sending one copy di rectly to Mr. Williamson.
I shal arrive in Chicago via Ozark Flight 736 at 5:40 p.m.
on Monday, 1 July, on ' I shall come di rectly to the Holiday Inn.
I expect that over dinner and af terward we can get our discussions taken care of in advance of the meeting on Tuesday.
I have to get a return f1ight at 5:00 p.m. on Tuesday ofternoon.
With best regards.
S i nce re ly, k
f LIMO ~
N. M. Newmark NMN:bj w Enclosure cc:
Mr. R. A. WI l iiamson p
<r gl s
v.
f 4643
4 h.0.
...... --el (
STRUCTURAL DESIGN CONSIDERATIONS AND SAFETY AGAINST EARTHQUAKES BODEGA BAY ATOMIC PARK UNIT NO. I by N. M. Newmark June 25, 1963 i
1.
INTRODUCTION This review is based on consideration of the material contained in Referen es 1 through 6, listed at the end of this report. In addi tion, an estimate has been made of the maximum credible earthquake intensity that may be experienced at the site, based on information f rom References 7 and 8.
Although the estimate used herein for the maximum credible intensity differs appreciably f rom that used in the application, (see aspecially Reference 3), if one takes into account the design stresses, reasonable values of energy absorption, etc., then for the appropriate
- choice of da. aping factors the net effect on the design obtained by the writer's analysis will not differ appreciably f rom the ef fect obtained using t ae procedures suggested by Dr. Housner in Reference 3.
- However, it is important to specify precisely the energy-absorption factor and the percentage of critical damping, as these affect the design parameters to a major degree.
Although it is entirely feasible to design the proposed reactor and the pertinent structures to resist the maximum credible earthquake intensity that might be expertanced at the site, such a design will 4643
1
, require careful attention to detail, and particularly to the possibility of large relative motions developing in the near vicinity, which must be j
taken into account in providing for power lines, pipe lines, and other means of communication beiween the reactor and points at or beyond the San Andreas f, ult.
It is a basic assumption of this report that the design will be made with methods consistent with the best available knowledge and information, including theoretical studies of earthquake resistant design,.
and will not be dependent solely on standard codes' and specifications developed for buildings of entirely different configuration, in which the hazards of partial failures are not nearly so severe.
2.
NATURE OF EARTHQUAKE MOTIONS In an earthquake _the ground-moves in a more or less random f ashion and in more or less random directions, although there is generally The a preferential di rection of motion parallel to a major f ault plane.-
vertical inotions are definitely less than the horizontal motions in Measurements have been regions such as those near the San Andreas_ fault.
made in recent decades by the U. S. Coast and Geodetic Survey of the
" strong motion" accelerations, as a-function of time, at a number of The record of strongest-points on the West Coast of the United States.
motion which has been obtained so f ar is that for the El Centro Earthquake of May 18, 1940. For this record, in the Nortn-South direction the
!, maximum ground acceleration was about 0.33 g; integration of the accelerogram gives_ values of maximum ground velocity of slightly less
}
than 14 In. per sec., and maximum ground displacement of the order of i
slightly more than 8 in.
On page 1 of Reference 8, it is,tated that "this is the most severe earthquake motion for which accurate records are now available; it may be considered as an earthquake to be expected in a specific Iccation in California wi th an estimated f requency of ~once in 50 year's, or more of ten if the r:gion is close _ to more than one active f ault.
Somewhat largermotions would no doubt. De experienced close to an epi ce n te r."
The ground velocity and ground displacement, obtained by f
integration of the ground acceleration record, are shown in Fig.1-1 on page 2 of Reference 8.
It is of interest to examine the characteristics of the motion shown by this figure.
ihe ground velocity-is characterized by many fewer peaks with a relatively longer time between successive f
peaks than the ground acceleration. The ground displacement has only a limited number of. peaks, with as much as five or six seconds between successive peaks of ground displacement. The total duration of large L
motions was of the order of 30 seconds for the El Centro Earthquake.
I In general, large motions occur in steps or pulses adding to the maximum value, but each step or pulse is characterized by
{
un having a relatively short time base.
Consequently, the maximum displace-is not at all a measure of the severity of the earthquake in ment terms of i ts ef fect on a structure which moves as a unit with the ground t
1 a on which it is supported.
On the other hand, the maximum ground velocity is a reasonable measure of the energy linparted to the structure, and consequently the energy that must be absorbed by deformation of the structure.
Hence, the maximum ground veloci ty is the best single measure of actual damaging or des tructive tendency in an earthquake.
For very brittle s tructures, the maximum acceleration of the ground is the best measure of the damaging tendency.
This is generally true for structures of any characteristics whatsoever where their fundamental pe riod of vibration is less than 0.1 or 0.2 sec.
The motions near the ground surf ace, or at the surf ace, on which a structure is founded arise f rom the large and violent displacements generally aloag a f ault plane or zone, wi th the major disturbance originating at some depth of the order of 10 to 20 miles below the surf ace of the earth.
Consequently, the accelerations and
- iult are not velocities neer the surf ace trace of the epicentrr' substantially larger than thus at some distance of the order of several miles away, although,the maximum permanent aisplacement might be substantially greater near the surf ace trace of the f ault along which the major motion has tu an place.
It is possible for a structure of relatively compact form, well tied together, to survive even by bridging across a surface f ault.
It will Se moved bodily and subjected to large dynamic forces. On the other hand, a structure which has separate supporting elements bridging across a f ault may be completely wrecked by the large movements of one side of the f ault relative to the other. Mere proximity to the f ault zone is not necessarily a measure of the damage potential to a
i structure, independent of the characteristics of the structural design.
3 INTENSITY OF 10CK TO BE CONSIDERED In Reference 7, the estimate is made that the maximum accelera-tion in the maximum credible earthquake at or-near the site would be of the order of about twice that measured in the El Centro Earthquake record previously discussed.
Earthquakes of -Intensity about equal
.to the El Centro record have occurred in the past, and have even exceeded i t (notably the 1906-San Franci sco Earthquake), and i t can be expected that this intensity will almost certainly be-exceeded -in the-future. Although it is difficult to estimate the strength of future earthquakes, there is evidence that it is possible-for an intensity of the order of IX, on the so-called Modifier' Hercalli Intensity scale, to occur in or near the region of Bodega Bay. A crude estimate, based.
on the probabilities of occurrence of earthquakes of'various intensities,
. leads to the conclusion that the maximum credible earthquake near a f ault may have-maximum ground accelerations, on sol.1 or sof t, loose rock, of the order of 0.5 to 0.7 g, maximum ground velocities of-the order of 30 In. per sec., and maximum displacements of several feet. On bedrock the accelerations would be higher, (in the range f rom 0.75 to 1.0 g),
but the displacements possibly lower with about the same maximum velocity.
The maximum displacements at or in the f ault zone may be as much 'as 5 to :10 feet,_ dropping rapidly,_ however, to values of the order of 3 to 5 feet at points ' definitely outside the f ault zone, and to less than 2 to 3 feet at distances of the order of a mile or more away. However,
6-the displacements are significant only insof ar-as relative motions between parts of a large structural complex are concerned.
I With reasonable accuracy, the intensity of expected maximum motion, in terms of accelerations and velocities, may be taken as about twice that for the North-South direction of the El Centro Earthquake of May 18, 1940.
4 DESIGN TO RESIST EARTHQUAKE MOTIONS AND SHOCK The response of a structure to earthquake shock motions is generally most conveniently determined by use of the " response spectrud'.
(See Chapter 1 of Reference 8.)
The response spectrum is a plot, agains t period of vibration or natural f requency of vibration,. of either the relative-displacements or strains in the structure, the relative velocities (measuring the energy absorption within the structure), or the maximum accelerations' of the components within the -structure, and can be used in ways that have been described in many technical papers and reports to determine the behavior of a structQre when subjected to an earthquake, or to determine the necessary strength of the structure to resist earthquake forces and moti ons.
Within the range of-periods of vibration of importance in most structures or structural element, the response spectrum can be approximated reasonably well by three constant limits or bounds. These bounds are generally described as the bound for relative displacement within the structure, the bound for relative pseudo-velocity (or circular natural f requency times relative displacement), and the bound for
4 i maximum acceleration of the masses of the structure.
For moderate amounts of. damping, of the order of 10 to 10 percent cri tical,- these bounds are as follows:
maximum relative displacement = 1.0 maxirum ground dir. placement maximum pseudo-veloci ty = 1.5 maximum ground veloci ty maximum acceleration = 2.0 maximum ground acceleration.
For relativel f small amounts of cbmping, of the order of 2 to 5 percent critical, these bounds are increased, to approximately the following values:
s maximum relative displacement = 1.33 maximum ground displacement maximum pseudo-veloci ty = 2.0 maximum ground veloci ty 4
maximum acceleration = 3.0 maximum ground acceleration.
For s tructures havica a f requency of v!bration in the funda-mental mode of more than 1.5 to 3 cycles per second, the acceleration bound is the appropriate design parameter.
For structures having c lower frequency, down to 0.1 cycle per second, the pseudo-velocity bound is
+
the appropriate design parameter.
Consequently, it would be only for extremely flexible and long pcciod structu es, having a period of more than 10 seconde, where the displacement bound would be of importance, and hence it can be neglected in the futher considerations of this report.
1 The use of the above,pectra in design involves a choice of the level of stress considered and the amount of plas tic deformation f
pe rmi ss i b le.
Design at wurb.ing stress rather than at yielding implies a f actor of safety of the order of about 2, using the working stresses 4
l In the " Uniform Building Code", as a basis, (not increased by the a
i f actor of 1/3 for lateral loading, however).
Hence a spectrum intended 5
to be used at yielding should be divided by a f actor of 2 in order to be consistent with a spectrum intended to be used at working stress levels.
i A reduction in the design r.pectrum is possible if one is i
willing to allow yielding in.the structure after an earthquake shock has occurred. The effect of inelastic be avior of structures is described f
in more detail in Chapters 1, 2 and 3 of Reference 8 However, in i
the following we shall be primarily concerned with structures oesigned-to remain elastic in order to provide for a sufficient margin of safety j
under unusual condi tions where partial f ailure or overstress might be extremely hazardous.
Since the basis for design described in Reference 4 involves design for normal working stresses, under the Uniform Building Code, in oroer for the recomrended spectra herein to be comperable,;the values described in Section 3 of this report must be divided by a f actor of 2.
l When this is done, the spectrum in Reference 4 for 2.5 percent damping, lies reasonably close to the values obtained f rom -the arguments herein, for structures founded on soll, but lies considerably below in.the range i
of f requencies greater tha n 2 cycles per second, for structures founded 07 rock. For such structures the spectrum value given in Reference 4 for 2.5 percent damping corresponds to a maximum accelera tion,. for design purposes, of approximately 0.9 g, whereas the recommendations given I
herein would be between 1.1 and '1.5 g.
4
.g.
However, this is not a matter that is impossible to resolve.
To increase the proposed design spectrum in the appropriate range of 2
frequencies wetid not involve undue hardship or inconvenience.
The provision for dif ferential motions within' a structure is more complex.
However, there are two f actors that must be considered.
i i
Where - dif ferent elements are founded on the same fi rm base but-i i
have connections between them, each of them can respond relative-to the base with a response determined f rom the shock response spectrum.
The relative displacement determined f rom the spectrum is the displacement f
of the mass of the structure using the base as a datum. - If the elements i
have relatively small damping, their relative displacement (between the individual masses) can be the sum of the absolute naxima of their displacements relative to the base, since whey may under certain conditions 5
deflect out of phase wi th one another.
Provision for such relattue displacement rust be made-where the responding elements are connected together in any way by connections, piping, wiring, etc.; This is true even if they are housed in rigid containers, but are f ree-to displace l-or deform within that container.
1 The second type of di f ferential displacement that-mus t be -
considered is that corresponding to the dif ferent base motions that may be experienced by the dif ferent parts of the base of a structure. Where a large structure is.not founded on a single integral base footing or mat, the individual parts can move relative' to one another, both under transient conditions, and permanently. The magnitudes of these relative
f motions are dif ficult to determine, but they should be provided for -
through-use of flexible connections, etc., so as to avoid tearing or i
damaging the vital parts of the structure because of the motions imposed on the different parts of the support.
This condition is not one for l
which it is impossible to make adequate provlsions in the design. However, it is not yet spelled out completely and in detail in the description of the proposed design.
Where flexible connections or piping with such connections 1-are used between di fferent parts or components in a complex structure, l
the joining or f astening to each of the parts must be considered to avoid tearing or rupture at these joints.
Fastenings of piping to the reactor shell, and other connections of various kinds, must be designed to provide for the requisite forces. These may be larger than those corresponding to the accelerations for a mechanical or structural element supported directly on a foundation which moves with the intensity i
of ground motion described above. The reason for the increase is that the responding element to which the piping, etc., is f astened, has a motion of its own which may be larger than that of the foundation or base, and may, in f act, be more or less periodic, giving rise to much larger forces in elements ported by it.
In other words, more sophisticated analyses of the sys tem, taking into account the fact that the system is a multi-degree of f reedom dynamical rystem, are required if the design is to avoid damage to component parts and elements connected to flexible or responding objects.
Provision must be made also for surges of-water or fluid that may act upon submerged elements or nearly submerged elements of the structure.
b a-s
J
.)
Such surges, caused by the earthquake motions,_ produce forces on the i
components which may ccuse damage if they are not-provided for in the design,
)
Finally, attention must be paid to the dynamic stability
~
of earth or rock slopes ir the vicinity where such slopes may, under the influence of earthquake. motions, slide down and out in such a way as to cause large and unusual forces to be applied to-structures,.
t piping, cables, etc., in the path of the motion. Provision can be made to increase the static f actor of safety of earth and rock slopes to the poir.t where they will not be subject to failure under the dynamic condi tions arising in an earthquake. The precise conditions are dependent on the properties of the materict and the intensity of the earthquake motions, but in general factors of >;fety against e
f sliding, under static conditions, of the order of 1.5 or more are required to prevent difficulties. Critical slopes, the f ailure of l
which might endanger either the structure or the utilities connected with the structure, should be examined to provide for an adequate-I f actor of safety under dynamic conditions.
l l-5.
RECOMMENDATIONS AND COMMENTS The condi tior.s at the site do not appear to be unduly or unreasonably hazardous nor are the motions to be expected unduly or unreasonably large compared with those which have occurred in other I
regions of the world or even ir the state of California. -Although the displacements and the veloci ties and accelerations may seem quite
~
i n
! 4 i
^
large, much larger shock motions are experienced in ships and 4
submarines, and materially-larger motions have been designed for in underground control centers and missile launching facilities, i
It is entirely feasible to design the proposed reactor to resist the maximur credible earthquake shock at the site.- To insure that the i
design will be adequate, however, ' requi res further. exarrination of a i
number of f actors in detail, which can best be done during the course i
of the design i tself.
A dynamic analysis of the final structure, as designed, s,ould be made to assure that adequate provisions have been made for relative motions, etc.
The provision in Reference 4, Part 3, I
that "in addition to the foregoing clastic design, a further analysis will be made to insure that. ground motion five times as intense as the design spectrum will be required to produce. incipient failure o#
1 structures",- is more than adequate to give a reasonable assurance 4
of appropriate strength for the structure, if the -analysis is adequately-j made. The analysis - should provide for. consideration of relative _ motions, however, as well as for s tresses wi thin the s tructure, i
Special attention should be paid to the ef fects of the-l carthquake accelerations on bri ttle and critical elements of the reactor itself, including the fuel rods and the_ control' rods. A!dynami c analysis of these parts of the structure,;taking into account the motions of the container in which they are placed, as exci ted by the earthquake motions. of the ground, should be made, i
s I
i i
w.
w
r
. 6 REFERENCES 1.
Preliminary Hazards Summary Report, Bodega Bay Atomic Park Unit Number.1, submi tted by Pacific Gas and Electric Company, Dec. 28, 1962, (Docket No. 50-205), See especially Chapter V, Section D.
2 Report on Earthquake Hazards at the Bodega Bay Power Plant Si te, by Don Tocher and William qualde, Appendix IV to Reference 1.
3.
Earthquake Hazards and Earthquake Resistant Design - Bodega j
Bay Power Plant Site, by George W. Housner, Appendix V to Reference -1.
4 Amendment No. 3 to Reference 1, Part III.
5.
Amendment No. I to Reference 1.
6.
Amendment No. 2 to Reference 1.
7.
Report by Frank Neumann.
8.
Design of Multistory Reinforced _ Concrete Buildings for Earthquake Motions, by John A. Blume,.Nathan M. Newmark, and Leo H. Corning, published by Portland Cement Association, Chicago, 1961.
.