ML20234A937
Text
{{#Wiki_filter:' j ,,,,~.. -,,-... -__ 2 g j 4
- y l
H O LM E S 8 N A R.V E R.. I N C. i EN GIN E E RS
- CON STRUCTORS eso south reournoA sTnzzT LOS ANGELES. CALIF. 90017 y
TELEpwont er7 4377 ] May 6,1964 Mr. Edson Case, Assistant Director Division of Reactor Licensing ) U. S. Atomic Energy Commission -{ Washington 25, D. C. t Dear Mr. Case During a discussion on February 28 and during a telephone conversation on March 16, you expressed a desire to obtain some data on two ques-i tions originating in the ACRS, which I understand to be essentially as follows: ) (1) To what extent do seismic considerations influence reactor costs ? j (2) How much seismic resistance is available in reactor facilities not designed t.o withstand seismic forces? Within the limits of the data at hand, I have attempted to provide, in the enclosures transmitted herewith, some qualitative answers to these que stions. I wish to reiterate my previous statements that to furnish more definitive data is a task of large proportions, which would still not remove the elements of conjecture. In my opinion, the results of j f. more extensive investigations would generally substantiate the estimates j { contained herein. I 1 Sincer ely, j 4 HOLMES 8r NARVER, INC. f WP9'++G a Robert A. Williamson RAW /es N0llying;p p ggggggg Enc: 2 ja gg;;;ggg ,NH33 AC' 3ll3 3l TlY 'S'n d 90 8 J g;.9 gq, J
- ^l?33'd l7assesy gg53 5-665 pg c
ti 3-j b bk S
l U ) The Effect of Seismic Considerations on Engineering and Construction Costs This materialis submitted in response to questions concerning the in-fluence of seismic considerations on the engineering and construction costs of reactors. It is emphasized that the almost complete lack of pre-existing data makes it impossible to provide precise answers and makes it inevitable that the conclusions offered are matters of opinion, i l The bias of opinion can be avoided only by making complete detailed cost estimates based on comparative seismic resistant and non-seismic de-signs of at least several nuclear power plants. The costs considered herein are the structural engineering costs and 1 1 construction costs. Structural engineering costs are placed in two cate-t 1 gories (a) process and equipment items (b) supporting and enclosing ] structures. Two seismic design approaches are considered: (a) the I conventional approach, which, for purposes of this discussion, is con-sidered to consist of the application of static lateral forces proportional to the tributary mass of the component being investigated (b) special analysis, which considers the lateral forces to be a function of the period 1 of vibration and damping of the component as well as its mass, Structural Engineerim Costs j Process and Equipment Items i To obtain a crude cost index, and as a matter of convenience, reference was made to a previous investigation which had been aimed at establishing the cost of computerized pipe stress analysis, including the usual the rmal and dead load analysis, and static and dynamic analysis for seismic forces. For the purpose of this discussion, it is assumed that the, cost per piece (defined as an elbow or run of pipe between elbows) for thermal stress analysis is $10. 00 and subsequent reruns $6. 00, and that analysis for lateral forces in two mutually perpendicular directions would be required. The data referred to indicates that when the lateral force analysis is of a 2551 i 3 0:
. - - - -..~ 1 ) I ] , conventional nature, it is reasonable to assume a cost of $3. 00 per ) piece per direction, making a total of $6. 00 per piece for the two j directions involved. This data also indicates that where the special seismic analysis is used, involving the first three modes, a price of f e $20.00 per piece per direction is reasonable, giving a total of $40,00 ) { per piece for the two directions. Using this data, it is estimated that consideration of seismic forces causes an increase in cost of pipe stress analysis in the range of I ( about 20 to 40% where the conventional analysis is used and an increase j in the range of 250 to 400% when the special analysis is used. The lower values in each case assume three thermal analyses for each two-directional seismic analysis. The upper values imply two thermal 1 I analyses for each two-directional seismic analysis. These htter in-creases are large. However, a better perspective is obtained when it is realized that, in the caise of the special analysis involving a total of ) i } 100 pieces, the total cost would be in the range of $20,000 including some allowance for reruns and processing of the input and output. This amount is probably not more than 10% of the total engineering cost for process and equipment items of a power reactor, but it is signifi-cantly higher than the cost of a conventional analysis. It may be possible to simplify the seismic analysis by supplying seismic bracing sufficient in amount to divide the system into components having natural frequencies which (1) are capable of being bracketed within f limits by simple means or (2) are achigh that a " rigid body" response applie s. In such cases there is more confidence in the seismic perfor-mance of the system because of its lesser complexity, and a manual seismic analysis may be more economical than a machine approach. It is reasonable to conclude that the special analysis is significantly more expen'sive than a conventional analysis. Whereas the additional cost of the conventional analysis might be ignored, the added cost of the special analysis would have to be considered in estimating engineering costs of, 9 g"r .I
1 .w l 1 () q _) ] l .the process systems. ) l Supporting and Enclosing Structures It should be noted that in the absence of seismic considerations a wind t i analysis of enclosing structures is still required, and that the conven-tional seismic analysis is similar to that used for wind forces. It is more complex, however, due to the need for calculating masses, and t centers of mass. In addition, whereas the total wind force acting on a building varies, depending on whether the wind impinges on the narrow or the broad face of the building, the design seismic force in l : the conventional approach is invariant with direction. l l Where high seismic resistance is required, the casual treatment of 1 joints and connections which is often found in wind design cannot be considered adequate. 'E provide an assured degree of structuralin-tegrity, the engineer must take on the added burden of extensively detailing joints and connections and must carefully avoid materials and configurations which behave in a brittle manner. This added task i causes an appreciable increase in the cost of the structural engineering. ) In the case of reactor structures, complication due to height, such as 1 in tall buildings, is not a factor; however, complex internal arrangement of floors and walls (as in the 105 N building of NPR, for example) may I cause difficult problems in estimating relative rigidities of the various l load paths. i Containment vessels of power reactors, directly supported on the ground, I in general, have a high inherent seismic resistance. and impose no diffi-cult problems except for local conditions such as anchorage at the base and conditions at penetrations. Reinforced concrete supporting structures for equipment located within the containment are massive and usually only require proper arrangement and appropriate minimum percentages of l reinforcing steel to be highly resistant to earthquake motions. Account-ing for the fact that the conventional seismic analysis is more complex l 1 i j
l -) ~ 4-C) )' ] i j o than a wind analysis and that detailing is more demanding; itis estimated f that the structural engineering costs would be increased in the order of 5 - j! 1 to 15% where the conventional seismic approach is used. In the special seismic analysis, which utilizes idealized damped spectra, the frequencies of vibration of the component being investigated become an important design parameter. The consideration.of higher modes, which may be required occasionally (as in tall stacks for example') can increase the structural engineering costs for these particular items. possibly as much as 100%. However, it is to be emphasized that the calculation of the period of vibration for the first mode only does not t usually involve comparable difficulties, and is a problem which structurai engineers frequently encounter in structural design under the current i Uniform Building Code. The resistance requirements resulting from applying the special analysis. ] can be as much as 600% greater than those associated with the conventional approach, when the spectra used are based on ground motions of intensity like that of the El Centro 1940 earthquake. Hence, in some instances, the i designer may be hard put to provide the required resistance without special measures. In general, because the forces are high compared to those of usual experience, the more casual practices which are justifiable j when seismic forces are low, tend to be replaced with more refined pro-cedures in attempts to minimize the effect of these large forces. The re- ] sulting increase in engineering manhours is not trivial. i Accounting for these factors, it is estimated that the cost of structural engineering for the enclosing and supporting structures of a nuclear re-actor facility might be increased by 20 to 50% as compared to that of a non-seismic design, when the special analysis is used. These increases are quite large, and would have to be reflected in the engineer's fee. It is evident that the effect on overall costs is much less when it is considered that the structural engineering allotment may be about 30% of the total t engineering budget and that the total engineering budget may be of the order of 15% of the total construction cost. Even if the structural b) 4 . E2C i E1 I -- - - - - - - - - -
[ Q ] l i ~ . engineering costs were doubled,_ this increase would represent only
- 4. 5% of the total construction cost.
Construction Costs ) While seismic considerations appreciably affect engineering costs,- their effect on construction cost is negligible in the case of most reactors when the conventional seismic analysis is used. This is known to be true in the case of the NPR reactor for example, where the design basis was comparable to that of California power plant practice. This practice utilizes a horizontal force factor of 20%g-- a value considerably higher than California building code require-j ments, but still considered by utility companies to be economically l justified. With a force factor in this range, the resultant increase in cost is likely to be much less than the range in bid prices. ( The increase in construction cost is, at least to a first approximation, proportional to the lateral forces applied to the structure and inversely proportional to the allowable stresses specified. The special seismic analysis, which utilizes spectra such as those currently proposed for Bodega, in conjunction with allowable stresses limited to normal working values, may result in a fairly appreciable increase in construc-tion costs as compared to that which would occur using conventional power plant criteria. In the case of the Bodega reactor, this increase is estimated to be in the range of 5 to 15% of the construction cost. i l.
,'..,.{ {} ) S'eismic Resistance Provided by Non-Seismic Construction 'This materialis in response to the question "How much seismic resis-tance is furnished by construction wherein seismic factors have not been considered? " (Referred to as non-seismic construction. ) Seismic resistance of such construction is highly variable and practically impossible to estimate with any precision in those cases where positive load paths for transmitting seismic forces have not been provided. Som e insight to this problem was obtained through an evaluation of the various reactor facilities investigated during the course of the writing of TID-7024, Nuclear Reactors and Earthquakes. (See, for example, Appendices A and I i B.) l j Further experience was provided by the subsequent review of the earth-I f quake resistance of the existing reactor facilities at the Hanford Works. 1 (Final Report, Earthquake Resistance of the Existing Hanford Production I i Reactors, Report HN-146, June 1960, Confidential, Restricted Data). l These investigations indicated that, in general, containment vessels which are directly supported on the ground have a high degree of seismic resistance, as do most supporting structures within the containment vessel. However, equipment items and related systems tend to be vul-i nerable. Those within the containment or other enclosures rarely re-ceive an analysis for wind forces. t Lack of anchorage and deliberate use of laterally flexible supports for t equipment items is quite usual. Such an approach minimizes thermal' expansion stresses, allowing relatively stress-free movement of equip-ment and connected piping. The resulting system tends to be complex in its dynamic behavior, although it cannot be said, as a general 11e, that the system is necessarily lacking in adequate seismic resistance. Other examples of potentially vulnerable features include: Control rooms housed in structures of unreinforced masonry; wind-braced towers supporting tanks containing the emergency water supply; and tall stacks, designed for wind only. .M }
- r<
) r. As a judgement based in part on the studies previously referred to, it is estimated that most non-seismic construction would not be damaged i by weak earthquakes having a maximum ground acceleration in the range of 5%g. It is probable that such construction, and especially features subject to brittle behavior, would receive damage in more or less degree in earthquake shocks in the range of the lower limit of strong ground motion--that is, in earthquakes having a maximum acceleration not less than about 10%g and a total duration of not less than about 10 seconds. The economic burden of pr oviding a moderate degree of seismic resis-t tance in a reactor facility to be located in a zone of low seismicity is I primarily added engineering cost. In such cases, refined approaches l l are not necessarily required to achieve a sufficient amount of seismic resistance, provided that the design details are carefully designed and I t l properly constructed to transmit the forces used in the design. ) i It is strongly recommended that no projected power react 7r facilities ) i for location in areas currently regarded as seismically inactic-be l t designed without complying with certain minimum standards for earth-l f quake resistance. t {~ l 1-1 l l l i l .i [- ..-..;,_-.=== 2) u.- = :. -
f, 0 / E. k. v* 4
- 'k
[.a e., ( a .fi, y I, s, -. .,,o FROMt t ' DATE OF DOCUttENTa DATE RECEWED NO.: Bobert h M g A. WE131ameen L4Jh C'* 2557 1 kh M hf N R* =M 'WLAngeles,_M, _Rg7 1 Weassmoure L w a.. gSPORT: OTHER: 1 O R IG.: CC: + _.. OTHER: g b g k.f [.' 4 + .e acnow escasosny O
- "*"""""a*-
O
- ^ ' ' ^ " * * ' " ' ' '
NO ACTION fESCESEARY ] CORAMENT CLAestF.s POSY OFFICE FILE CODE: SYs e WQ "e W s REG. NO: ') P. "" '..y.. DasCRifrT8CNr Odves De Usciass1hed) /kttantoTO 14r. tramand%ing sheets Armishing DATE l k atCEfvED av l DATE x same ens to what estems de sesamis / E.wse 542 epamiderations Amr1===== roerter easte i w--l .l end 3) how mask estamisp resistesse is ,~Cuo.ua r., l ara 11ahle in remeter e emi== art do.- g [
- i stamed to withstand asiaste throne J
~ A (ukish Mr. Seas empressed a desire to r ebtain due to their artatastima as i _ ' ' [ g, s b::.b h i'
- J.-* l%,.,. ]g i
r, F
- e sesAnsts,
~ h' kN W Mh%g L.j r : 84 54-4k smettas sa missess. p.: ! w v31 r.,._ 2r,A, b ^ U. S. MOBHC SMBBSW COMMISSION 2 M., e u.....m o. T m.,..q,g, MAIL COkrKOL FORM roga ..-:.~. m:. -w n W,:TTad"% %..= ---.. + w. m.w..nwu; ::n;y-,;.:~M Y '[g Y@ j Mhg _ lsf*.;*'* J"DM %f I MC.d]3.= ffd=w"%g ?,'"m';*M '.;;.t"'*y/,~*G'.~~ 'h,*"9?"3' ;'.%gr4 =W-. yg*.3 r.gf .,v stc-s.3f.%.ryvy p m.? ~ -h..Cy1QQ;;r.: W: a.. ,...-.== =. -- .: 3m.=. ^ ^2.A~, w g.WW2W M ' *V .7:.~ys m.; TL.~} W a, a i %.* '** ;.,{A* +. e. l" &,,.-f. *,*-'M .yfA.Uw*I',,.& %,. W'..' "a.s.,. Y** u""5,A. s- *((l..s i ? Ee. ' ,.
- a _3 c, m.~..D,...._ w _e,.
.*.W'M 3 N ".,,.-. n~, ".*J.w,.c %. %.,.*;,'~., 4'
- , M,,...Q M %g. %[,.w O 7.,.u.*.**.*;',.,,,.
m- ~ e d en .w. e w"-s... ~e--+
- c;%.-
e m. .. e.. l w .r 9I*E b.. ---< u..#.-r. % :.:.p,,,,,.ff m. ~ m.-. M..- e.4ff _,,, re,u . +.=.,=,.%. = adam.',**~.++ M *J m y c.. ,, -, r._, ' A.M y._, .ugn +* -',*i s
- w -.. ~A a
4,. ~. .w,*, ;,.. + .. ~, -
- -..,.A.,.*~
1 .. +.. g,. me .~ . ~,, c :.. n. =;* s ..e a . w.m f .) ~ ^~ s?. uw.. ~,~g.. . m. m.._ n._ = m ; g w w w m.s,_s.n: w..~.,,;.,1.:.=.w w =.7. _c a-r =- n.. ,. y. ~. ., _. _ = 4, .,, n N' .,- +~' * * - , - _ -m".w1*","*'*3,., %.T... : - * : ', 'c _. C'."c -J',C._* ;.%..,.,4 1* " ;.: e "'A.*
- 9., .,. v 4..r - a
"~# --d.A~~~~- r-
- se 7 ;.;- _
[ _.s.i.'*** , +8 / f U[... M**** . - * ' * * * **.+ r, r .~**U 't .,,,j. - -e ..e..,; _ n. . * ' * + ~ -' ?- ...__'***A =~" :e *-~ :l' *-* h.% 9 ' ~* * *- -, _ - -
- #4'#*'**
D i, g '-8 b .p. g ' g*.ig. ew"/_
- U
W mL.- A 'k. - 4 .F r <-.* 1 l 4 i .f i e 4 4 . '. ?_ .._____m. _ _ _ _ _ _ _ _ _ - - - --}}