ML20244D969

From kanterella
Revision as of 19:15, 22 January 2021 by StriderTol (talk | contribs) (StriderTol Bot insert)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
Provides Final Rept Prepared W/Wj Hall Re Seismic Design of Plant,W/Assistance of Wh Walker.Rept Based on Fsar,On Ref to Certain Sections in Preliminary Hazard Summary Rept & on Applicant Presentations During 680710-11 Site Visit
ML20244D969
Person / Time
Site: Nine Mile Point Constellation icon.png
Issue date: 10/23/1968
From: Newmark N
NATHAN M. NEWMARK CONSULTING ENGINEERING SERVICES
To: Morris P
US ATOMIC ENERGY COMMISSION (AEC)
Shared Package
ML17055E652 List:
References
FOIA-89-101, FOIA-89-114 NUDOCS 8904240229
Download: ML20244D969 (10)
v • d • e


Text

- _ _ _

NATHAN M NEWMARK I co s 9u trir m r N olN c E nit s a ,s te vic Es, 11:1 CWil. ENGINEERING P L4 LCIN G s AC3S Cec =ittee Members

) '

SHb # HO:.t_

G HE - AA0 t ..

  • id A3""

WLF- - DO - '. ':

SHH -

1:n- u 4 F!

JMH - I 3

CPS t- iO l4 HSI v LS - E H3M Was c- 72.G

. ' ;! E L
r . Peter A. Morris,Iirector D~~

C;E v E5i ' 4 l'ivision of Peac tor i icons i ncDate Hailed /k[#/

l' , '. Atomic i r,< ryy Conwit s iori Us' ~

f

/ / M a' f -

W.e h i no t cin ,

i.C. 27545 9 * .

Re Contract A f (h9-$ )-2 667 Nine Mile Point f.ucleer Station,

^

[Y ]

Niacara Mohawk Ci,r:> ore t ion

~ l /

AEC Docket No.91-229  :.

7

Dear Dr. Morris:

p j

)

The folIowina is our final report, i;r epared by Dr . W. J. Hei1 l 9

and mys e l f , on t he sc ismic des ion of t tir hine Milc 'oint bucina f

Stat ion, with the ass is tance of Dr. W. P. Walker. This report is based on l

the Fina: Safet y Anal ys is Report (FSAR), on reference to certain <cctiens in the Preliminary Hazard Sumn.ary Report (PHSfs) wh ich s upr> l ied s ome de t a i led information not available in t he FSAR, and on presentat ions by t he applicant during our visit to the site on 10 and 1I .lu l y 1 %8.

In order to facilitate our rev iew, t he a.ipl icant t ransmit ted to us inf ormall y copies of t he br ief ing char t s used in his t resent at ion dur ing the '

site visit, and copies of additional calculations on which the briefinq charts 4 J

were based, as well as computer outputs for the piping dynamic analysis. We v+

t~

are especially appreciative of the applicant's cooperation and willingness /

  • )

to supply information regarding the details of the procedures used by his e ng i nee r s . In our opinion the applicant h.is voluntar il y cons idered many l

enginees ing ques t ions involved in the des ign and has res ponded fully to the var ious ques t ions raised by vs. Since we did not have any contact with t his

,u .

-- 3

.y X lO ~? -

.  ; 9,n m \v[

\

Q )

8904240229 89041b P&LJF9fe

'J '

JJ .

.. , ' plant during Ihe construction permit stages, it would have proved impossible

/ ' to have determined the bas is f or certain aspects of the des ign without the complete and willing cooperat ion of the applicant in f urnishing detailed -

J information and in making additional studies in response to the ques t ions iL raised informtlly during our meet ings.

7, r.eneral Description 5

.j'

  • The Nine Mile Point Nuclear Stat ion contains a boiling water reartor
  • l4 with a pressure suppress ion f ys tem involving a dry well and suppress ion chamber 1

i - f or pr imary contai nment , with secondary containment provided by a reactor building ventilated to a stack. The des ign capacit y of the plant is 1,538 MWt (500 MWe ne t ) .

i The plant is located on the southeast shore of Lake Ontario, Oswego Count y, New York, 36 miles NNW of Syracuse, and f ive miles NF' of Oswego, t

.p New York. The geology at the site is characterized by overburden of J

( approximately 11 ft. in thickness overlying light grey to greenish grey

.y

.} ' moderately hard Oswego sands tone. The maj or nuclear s t ruc t ures are f ounded on

+

bedrock.

1

The plant is designed for a seismic input corresponding to a maximum 1

j hypothetical earthquake of 0.119 horitnntal ground acceleration. This design basis antedates the current use of an Operat ing Das is f arthquake and a Des ign

{~

Bas is f'ar t hquak e , wi t h t he latter generally twice the former in i nt ens i t y, Because of this c ircums tance, cons iderat ion was given in our review to the rl capability of the plant f or safe shutdown, cons ider ing the conservat ive des ign

s f bas is that was generally used to account for the single earthquake considered.

-F Seismic Response Spectra lhe f ollowing comments are based on the conf erences held with the '

applicant in July 1968, and on the information in Supp lement ? t o t he F '.AF t o

/

m

( - - - - . - . - -

_y_

s upplement .his answers to our ques t ions , and supplemental data including

/- '

computer out puts of c alculat ions , s ke tches and drawings of the _ s im li f ied l

s t ructures cons idered ' in the analyses, and some descr ipt ions of the' output in terms of equivalent' accelerat ions, etc. , presented at t he conf erance.

The earthquake cons idered for. the design of the plant was ore

.. corresponding to a maximum ground acceleration of 0.119, but with a response

'l. .

f spectrum that is not-of the normal type commonly used now for other plants.

I The complex response spectrum shown in Page C-22 of the f irst supplement to i

the i H5R, ref erred to in the FSAR as the one used in the analyses , was not

, the one actually used in the analysrs cons idered herein. The actual response

'f I spectrum used has r'ot been furnished to us. However, a briefing chart t

., showing the general nature of the resporee spec t rum apparent l y used . was made available, and the descr ipt ion of t he computer solut ions gi ve i

! data f rom which the response spectra actually used can be inferred. From a cons iderat ion of these data it appears that in the region of major interest f the response spectrum used in most of the calculat lons can be approximated e

-i by a sweeping curve having a velocity response of the order of about 3 in/sec.

f j at 0.5 Hertz, rising to a maximum of about 4 in/sec. at about 1.5 to 2 Hertz

} and dronning of f to 3 in./sec. at about 5 Hertz, where the curve becomes tangent l

- 1* to a line corresponding to a constant acceleration response of the order of

.i

[

about 0.2 to 0.59,which drops off at much h igher f requenc ies,over about t

f 40 He rt z, to about 0.119. Some variations in these values are used to correspond to various values of damping, but t he variat ions are not s ubs tant ia l .

1 '

For moderate amounts of damping, of the order of 2.5 to 5 percent, this 5' response spect rum is in reasonably close agreement with t hat given in TfD 7024 for f requencies higher than about 2.5 Hert z, but departs by success ivel y I

l

- 1 i

, n

.. y 4

/

9reater amounts' f rom the TID response spectrum for lower f requencies. At about i Hertz there is a f actor of dif f erence of nearly 2, and at about

. 0.5 Hertz about 3 In other words, for moderate amounts of damping, the l' response spectrum used by the applicant would in general be unconservative -

i l below a f requency of 2 Hertz, with the lack of conservat Ism corresponding

}, to a factor of 2 at 1.0 He r t z and a f ac t or o f 3 a t 0. 5 He rt z .

'l For the s tack, for which the detailed calculations are poss ible to s

fl check in some detail, the response spectrum used in the computer computat ions corresponds to a velocity response of the order, on the average, of about 3.0 t

to 3.5 in./sec. up to a f requency of the order of about 4 Hertz and then

! -j

i corresponds to a cons tant accelerat ion response of the order of about 0.20 to

[ 0.229 above that f requency.

I In our oninion, a more appropriate response spectr.* tor the stack,

( assuming the applicant 's value of 7.5 percent critical damping, would be one j that corresponds approximately to one-third the conventionalind El Centro response spect rum, which would correspond to a cons tant velocit y response of I

f about 7 in./sec. for all frequencies below about 2.5 Hertz, (which is 2.0 to 1

'! 2.3 t imes the applicant's values) and a c ons tant accelerat ion response of 0.229

.l above that frequency (which is about equal to the applicant 's values),

1-

[ Similar, but not exactly the same, corrections apply for other

,.. degrees of damping for the spectra that should be used for other parts of the 4

s ys t em. It is necessary to keep these comparisons in mind in interpreting the capability of the various parts of the plant.

Ano t he r wa y of s umma r i z i ng t he response spectrum actually used in

^{

[* the calculation is to point out that for relatively stiff or rigid i t ems , and l

l L in general for most items of equipment and instruments, where the frequencies

! are over 2.5 Hertz, the response spectrum used does indeed correspond to that T

! for a maximum ground acceleration of 0.119 However, for large buildings,

c. .

-)

5 s tacks , long runs of unsupported pipe, and in general for all items with a b lower f undamental natural f requency, t he response spect rum actually used corresponds to approximately 0.069 maximum ground acceleration.

Int erpretat ton of Applicant 's Seismic ites Ian Cri terla

'In general, the des ign was made. for .the earthquake specif ied by the applicant, at working stresses or i n s ome instances with an augmentation of i ,

i*

one-third over the normal working stresses. In general, for Class I items t

t he normal wor k ing s t resses we re used wi t hout the increase. For the turbine i-building the one-third increase over the normal working s t resses was used but the actual stresses cre stated to be generally within code allowables.

, For Class I s ys tems in the turbine building the one-third increase was not applied.

For s tceI and concrete and for elements made of.combinat ions of these materials in general, yielding occurs at stresses of the order of about

] twice the working s t resses, or at a s tress of the order of about 1.5 times

(

the working s tresses increased by t he one-third augmentat ion f ac tor. Hence,

!. for Class I s tructures and s tructural elements and for equipment that is i

pr imarily s t ruc tural in its characteris t ics , where the f requency is higher

'I than 2.5 Hertz, the design is generally adequate without yielding for an i

i earthquake hazard of the order of 0.2 9. However, for longer or more flexible i

,. I t ems or in general for structures and structural elements with a f requency J

I below 2.5 Hertz, the capability in terms of res istance to yielding is of the b

{ order of 0.119 because of the inadequacy of the response spec t rum used in the low f requency range, subject to an uppe r revis ion, however, because of the conservat ism of the methods of analys is generally used by the applicant , as discussed later.

For vert ical earthquake exc itat ion, a "s tat ic" vert ical accelerat ion factor was used corresponding to one-half the horizontal acceleration at the particular point in the building. These vertical forces were applied to the structure and stresses computed from them.

g- -

c ,

. 4 For .the pressure suppression chamber or torus a hor izontal stet ic

/- coefficient of 0.15g was applled, combined with a vert ical coef ficient of 0.0559 _

Se lsmic De's lan and Ana l ys is -

In general the appile ont's analysis corresponded to the calculations

'i of the maximum accelerations in each mode, the combination of the nodal accelerations as the square root of the sums of the squares of the nodal accelerat lons cons idered, in most ins tances us ing 5 modes of respor.se, and f the determir.at ion of the forces corresponding to these accelerat ions as s tat ic

.! t

-; forces f rom which momenti and shears in t he s t ruct ere were de termined. In the case of the stack, the turbine building, and the reactor building, an (

l l

addit ional " external" input was cons idered as if it were a separate mode, j

corresponding to a horizontas acceleration of 0.119 at the bsse and varying v.

J linearly to 0 at the top of the structure.

I ihis is admittedly an incorrect me t hod o f a na l ys i s . An appropr iate

( correct method would involve the computation for each mode of the moments

,e

,_ and shears, or accelerat ions and displacements or other modal responses, and 1

[ the conbinat ion of these modal responses, taking into account the excitation of each mode as determined f. am the appropr iate response spect rum curve, i

t by combining the maximum modal responses us ing the square root of the sums

-l of the squares of the individual responses cons idered. The combining of the t

'f accelerations and the determinat ion of scismic forces f rom them gives forces i

{ which correspond to over-es t imates of the octual moments and shears in the lower parts of t he var ious s t ruct ures by f actors which may range f rom a few g percent for moderate to high f requency items, to as much as a f actor of 1.5

.V l

,1 for low f requency items. Hence, the combination of the various degrees of 1-conservat ism used in the individual stages of the analytif leads generally to a capability on the verge of yielding of the order of about 9.15g to 0.2g f

4

l .' -

-7 I

for structural components and for other items for which stress is the crimary L

l '

criterion.

A part icular example was analyred 'n detail. Th i s is the stack, which has a f undament al period of 1.90 secs, and which was analyzed f or 7.5 j percent damping, lhe shears near the base were computed by us f rom the i l

3 applicant. 's nodal anal yses , us ing the values given by him for the model

't 6 respontes and his response spectrum, for the lover elevations in the stock 1 the square root of the sums of the squares of the modal responses of base r

modal thears corresponds to 4.5 to 5 percent of tiie we ight above t he point considered. This 's in cont rast to the values computed by means of the I applicant's procedure for the same cend i t ions and input s , of 11 to 13 percent considering the first f ive modes onl y tiut without the so-called " external" I

g additional t r iangula r mode. With the latter included, t he appl icant 's I calculat ions , as summar ized by our procedure, corres. pond t o 14 to 14.5 percent r

(

s of the weight. The applicant 's values are reported in the briefing charts /

I i as 14.5 to 15.2 percent .

I 1

. i An interpretation of these results is in order. With the appropriate 3 spectrum for an 0.I19 na ximum ground occ"lera t ion, as i nd ic a t ed ea r 1 it r., we would es t irw e resnons e values 2.0 t o ?.; t imes as high as the values f

referred to above, of 4.5 to 5 percent of the weight. Consequently, the

{*

base shears that we would compute as being appt roriate f or an 0.119 i

excitation would be 10 oercent to 12 percent of the weight of the stack.

j The value for the base shear used in the design by the applicant was i

l presumably 14.5 percent of the weight of the ack. He nc e , this would correspond to a capability, based on our concept of the appropriate shape of spectrum, of 0.139 to 0.159 maximum ground acceleration.

In the earthquake anal ys is 36,000 ps i st ress was used for the

(

re inf orcerrent , which would ind ica te sonie add i t ional marg in t r ior to yielding,

.s '

c i .-

l

.and hence a capability before yielding or collapse of the order of . os s ib l y l~ l f 0.16. to 0.1P9. makirnum ground accelerat ion.

F The overturning moment at the base of the stack is also' conservatively-

-figured b"y the applicant; in~ general the stack has a capability i

1. cons iderably greater than the earthquake for'which it was des igned, even

,.) allowing.for the inadequacies in the response spectrum used in the design.

':i

,jf

.In our opinion,_ in general the various items of equipment, 6

ins t rumentat ion, and mount ing racks eppear to have a capability of the order of S. I5g or' more, because of the conservat ism in the methods used in the ana l ys i s , in spite of the fact that the response spectra may not have been entirely appropriate.

+

F ip i no-The piping des ign for the plant was made in accordance tvith the C

I 1955 Piping code, Bl.1, and the applicant has not yet completed analyses of

(

the capability of the piping design. The applicant 's addi t ional analyses i

j 'which are being made will be based on the use of horizontal accelerat ions 1

computed at the various elevat ions f rom the analyses described above, with

  • vert ical accelerat ions one-hal f the hor izontal accelerat ions at each point.

i j

These values are stated to range from n.llg at the 198 ft. elevation, near 14-the base of the building, to 0.389 at the 340 ft, elevation, for the horizontal f accelerat ion coef f ic ients.

g From the results of these analyses the applicant

[

j has s tated that he will add supports and s tays as required to increase the capability of the piping to be in reasonable agreement with current codes such J. as USAS B31.1.0 j

A descript ion of the method of analys is used by the applicant was made available to us. The applicant describes the piping s ys tem as idealized as a mult i-lumped mass sys tem interconnected by weightless I

1 i

- - -- -- - - - - - - . - -- a

.. _ . _ ._ u ._

s, o.

_l e*

s. <

9..

structural elements. These s t ruc tural element s take into consideration I<

flexural, shear and axial defornetion. The maximum accelerat ions in each of therfirst five modes is determined and the individual modal accelerations

,[ are combined by taking the square root of the sums of the squares to obtain net. acceleration values.

The seismic f orces act ing on the system are determined by multiplying 1

i. ' t he mass a t cach node point by the acceleration at that point and then the

.V O. y . s tat ic forces 'are used to obtain moments , shears, and deflect ions for the i

p iping s ys tem.

In orderfto take account of the fact that the bends in the piping

[ j were treated as square corners intersecting at the bend working point, the t ,

res ult ing~ s t resses' were mul t iplied by a f actor of 2.

This Is not an accurate calculation and in general would not be

{.

1 .

j considered adequate. However, the method does have built-in conservat ism 1 as does the method used for other parts of the facility as described earlier.

h.

'k':

Hence, we are prepared to accept this bas is of calculation as being reasonable

.a.

17 for the plant, in view of the original des ign criteria, although we would not

.I be willing to cons ider such a procedure acceptable ab initio in a new plant.

{ Of particular concern to us would be the problem of relative defernations between points that are connected together and which may move x

l' ', ' in dif f erent ways. For exanple, the piping connecting the reactor vessel

[ through the drywell to the turbines is subject to differential motions at its two ends, both f rom the long period relat ive displacements and f rom the t'

t rans lent responses of the points of attachment of the piping. Consideration' i does not appear to have been given to this point, but it is likel y that the stresses arising f rom this cause are not large in view of the rock foundat ions, and the relatively small earthquake input under cons iderat ion.

.. e

i l

l Ifferts if l'ir nadoes q l

tnrnado loading was not considered in the or1 9 i nal des ign cr i ter ia.

[ i From the discuss ion in Sur plement 2 to the FSAR, p. 1-33, 34, it appears that

the plant can resist most of the effects of a tornado, but not missiles, etc. lhe applicant indicated that he will give cons iderat ion to the ef f ects of tornadoes and of high intens ity winds on the s tack and on ot her exposed L

} portiens of the facility and will take measures to avoid accidents following i

f a tornado that might impair the capabilities of the plant for safe shutdown.

i

Jns t rowntet ion and Cont roll the control room components are des ignated as Cla I items and the se ismic anal ys is used corresponded to an applicat ion of accelerat ion f actors I

i of A.?q harirontal1y, and O.Ir verticalIy as static loads. Ihe applicant has

) i nd ica t ed. t ha t provis ions will be made to insure the safe shutdown capability k

of the instrumentation and controls in ihe cont rol room under either seismic

( or tornadic conditions.

.- E umma r y

,. As a result of our review of the des ign provis ions and tha construction t

program to date, and after conferences with the applicant and his representatives and study of material transmitted to os through your office, it is our i

f conc l us inn that the Nine Mile Point Nuc. l ea r S t a t ion wi l l have a capabilit y V

j of res is t ing an ear t hquake of maximum ground accelerat ion of the order of i f 0.159, in general, in the sense that it will have the capability of safe Shutdown after such exposure.

Ve ry t rul y yours ,

/ f .')).! }' .

f'T N. M. Newma r k bj w cc: W. .l . HalI

['

s J. i. HaItiwanger W. H. Walker l _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __i

Retrieved from "https://kanterella.com/w/index.php?title=ML20244D969&oldid=3106178"