Text

Cable Tray Support Qualification Program
	ML20141B832
Person / Time
Site:	Seabrook
Issue date:	03/31/1986
From:	; PUBLIC SERVICE CO. OF NEW HAMPSHIRE
To:	;
Shared Package
ML20141B827	List: ML20141B823; ML20141B832;
References
	NUDOCS 8604070124
	Download: ML20141B832 (150)
	v • d • e

{{#Wiki_filter:_ _ _ - ATTAC11 MENT 2 i Seabrook Station i New Hampshire Yankee Cable Tray Support l Qualification Program l l l v ) 4F i i sisss e, l MARCH 31, 1986 1 i i I ReR "I8an 8t888 6 i A PDR I l

TABLE OF CONTENTS EA&R 1.0 EXECUTIVE

SUMMARY

I 2.0 PROGRAM OBJECTIVES AND SC0PE.................................... 2 2.1 Background................................................ 2.2 Obj ec t i ve................................................. 2.3 5 cope..................................................... 3.0 QUALIFICATION PR0 CRAM........................................... 4 3.1 Testing Program........................................... 4 3.2 Cable Tray support Qualification.......................... 6 3.2.1 Testing................................... 3.2.2 Combination Test and Analysis...................... 3.2.3 Analysis........................................... 3.3 Ot he r I s su e s.............................................. 7 3.4 Past Performance of Cable Trays in seismic Events......... 8 4.0 TEST PROGRAM DEVELOPMENT........................................ 9 4.1 Test Program.............................................. 9 4.1.1 Dynamic Test Program............................... 9 4.1.2 Development of the Dynamic Test Configurations..... 10 4.1.3 Connection Tests................................... 12 4.1.4 Coordination with Bechtel Raceway Support Program............................................ 13 4.2 Program Results........................................... 17 4.2.1 Dynamic Test Results............................... 17 4.2.2 Connection Test Results............................ 19 4.2.3 Correlation Analysis of Test Samples............... 19 4.2.4 Analytical Configurations.......................... 22 5.0 SUPPORT QUALIFICATION CRITERIA.................................. 24 5.1 Criteria Summary.......................................... 24 5.2 scope..................................................... 25 5.3 Evaluation by Test........................................ 26 5.4 Evaluation by Combination Test and Analysis............... 21 5.4.1 Primary Connections................................ 28 5.4.2 Brace connections.................................. 29 5.4.3 Nodeling Methodology / Evaluation.................... 29 1 - -. _ _ _ _ _ - - - _ _ _ _ _ _ _ _

~. IM.t.: or CouTsurs (Continued) Pese 5.5 Required Review Deta....................................... 32 5.5.1 Loads.............................................. 32 5.5.2 Desping............................................ 33 6.6 3UPPORT EVALUATION PROCEDURES.................................... 34

7.0 REFERENCES

48 APPENDIX............................................................... -111-

1.0 EIgCUTIVE SUIstARY Dynamic test programs conducted throughout the industry have demonstrated that cable Tray Support Systems exhibit significant seismic capacity. These dynamic studies have enabled the Seabrook project to pursue a test based qualification program. The program is based upon plant-specific te{tstojustifythesolemicqualificationoftheexistingSeabrookcabletray support configurations in their present state of completion. In the absence of this test based qualification program, a substantial amount of seismic bracing, and hardware improvements would be necessary to complete the original design concept. Cable tray support performance and behavior for typical Seabrook systems details and materials were evaluated in the laboratory during the initial phase of this program. This included full scale dynamic testing on representative support and tray configurations at high levels of seismic input. These proof and fragility tests were fundamentul to the development of a revised analytical approach for the Seabrook cable tray support qualification. The testing program demonstrates that the typical Cable Tray Systems exhibit substantial seismic capacity. The testing results have shown: (1) existing connections exhibit substantial rotational resistance and (2) the Cable Tray Systems exhibit highly damped response. The test data has been used to develop an analytical approach based upon actual and prtfictable system behavior. The final phase of the program ecuples this technology with the Seabrook installation by performing an individual support seismic evaluation i and component interaction review. This phase of the program provides the final documentation for the cable tray support qualification. It is concluded, therefore, that implementation of this program fulfL11s project commitments for the quellfication of the existing Seabrook cable tray support installation. i 1-l -e c-

rw--

w ,-------,,w-+ -,, - - - -, - - - - + y -,w_

2.0 PROGRAM OBJgCTIVES AND SCOPE 2.1 background seismic qualification of cable tray supports in certain Seismic Category I buildings is to be completed by utilizing dyn mic testing and and,1ytical methods to refine the existing project analytical methods. The decision to redirect the Seabrook cable tray qualification was influenced by the following factors: o Past shake table testing of Cable Tray Support Systems have shown that Cable Tray Systems exhibit inherent seismic capacity with varying amounts of seismic bracing. o Existing project analysis modeling methods of Cable Tray Systems could be refined to be more reflective of actual system behavior, o Dynamic testing is an accepted seismic qualification method by the USNRC Standard Review Plan (Section 3.10). 2.2 Objective The objective of the refined qualification program is to produce and implement a methodology which will optimize the project's use of available 4 resources while meeting appropriate acceptance criteria and margins. 2.3 Scope seismic qualification of cable tray supports by dynamic testing / analytical methods will pertain to supports located in seven of a total of eleven Seismic Category I buildings which still require final documentation of their qualification. In t.he remainh.g four Seismic Category I buildings, the design, installation and final documentation of cable tray supports based on the original design concept is complete. Figure 2.1 and the lists below identify the buildings with applicable seismic qualification programs. 2

.. _.... _ ~ -.... _ _... _ _. _.. _ _... _ _ _ _.. _. _. Seismic ouellfication Buildinz Status Buildinas Utilisina Refined Froaram f.po leted Buildinss control Building Service Water Punohouse toastor containment Building Cooling Tower (Unit 1 side) Primary Auxiliary Building Diesel Generator Buildir.g Containment Enclosure Ventilation Area Fuel Storage Building ENE Spray Equipment Vault Steam and Feedwater Pipe Chase (East) Electrical Tunnels Trains A and B 3

t - Service Water Pumphouse ,{ (completed) e --- Steam C Feedwater I ^y Pipe Chase (East) (not completed) O Reactor Containment Building ] 4 (not completed) Turbine l j Fuel Storage Suilding Suilding (completed) ~i

Containment Enclosure n

'1 Ventilation Area H, i, (not completed) .r- - -Prim.ary Auxili.1ry Building j (not completed) / N Admann.strat!O" t Building RilR Spray Equipment a Vault (not completed) J / ,/ I t.'ilu t I d l *' b !!jesel Hui li!i n.: i;ene r a t o r Cooling rowcr - ~ (not completed) Hu i lili n,: (Unit 1) (completed) (completed) I 2 --- a } STATl'S 4 i'.. up.l: ik n wl'"i'aiki sit:Al,II ICATio.'. i l e.fl6ti, l.l

3.0 QUALIFICATION pB0GRAN Isylsesentation of the qualification program will be performed on an individual support basis. A review of all supports is planned using as-constructed drawings and walkdown results. Individual support quellfication will be accomplished by implementation of one of the following th h (3) methods: (a) testing (b) analysis or (c) combination test plus analysis. Support qualification by testing will be utilized when the support geometry / hardware mass and seismic environment, etc., are bounded by a specific test case. Supports not directly bounded by the geometry of the test cases but whose behavior is consistent with the behavior of the test cases, will be qualified by a combination of test and analysis. Supports not represented by test models will be qualified by a unique analysis. An overall description of the qualification program is presented below. Detailed descriptions of the qualification methods are presented in section 4. Table 3-1 sunenarizes the implementation phase of the qualification program. 3.1 Testing program The qualification test data base was developed by shake table testing of four representative support configurations which are predominant throughout the seven seismic Category 1 buildings. The system behavior of each of the configurations was obtained by testing full scale systems. Figures 3-1 through 3-4 show standard consttvetion details of the supports simulated by test. Figures 3-5 through 3-12 show typical cable tray layouts which are conservatively enveloped by the test configuration. The test program data provides the necessary data to extract an understanding of system performance. The data is then used as an effective link to the analytical applications program as follows: 1. Connection test data is used to (a) establish appropriate spring rates to be inserted into support finite element models and to (b) establish appropriate connection rotatio,1a1 and stress limits. 1 4

2. Finite element models of the test case samples derived ith Item 1, are then compared to the system test data. Correlation of these analyses with the test results in tems of dynamic properties and response levels serves to verify the modeling approach. 3. The connection performance data developed in Item 1 is then ,e reviewed versus the system data. Utilizing these connection performance data will result in responses that are limited to less than those observed during system testing. This will ensure that supports qualified analytically and in conformance with the acceptance criteria will be within the response levels generated during system tests. Joint rotation, for example will always be limited to values less than those observed during system testing. Supplemental analytical evaluations a: e used to illustrate the severity of the test conditions and their resulting broad applicability. This is because the input margins included in the test programs resulted in bounding response data. In plant applications, the connection moments and forces, connection rotations, etc., are limited to values which are less than those recorded during the tests. Therefore, although the test samples do not j geometrically envelope 100% of the plant, they were subjected to responses which, by the implementation of our qualification criteria, will envelope the actual support system responses at Seabrook. A support-by-support review will verify that indeed the tested supports envelope the response conditions of,the vast majority of the existing support installations. This confidence is due to our parametric surveys which have yielded the following data: o The Test Response Spectra (TRS) exceeds the horizontal ARS of all elevations in the control Building by a mintanam of 60% at all frequencies above 1 Hz. o Fifty percent (50%) of all trays in the Control Building are loaded to 20% or less fill by area (notes 40% fill was used in the test program)... -.

e._u

+

.e .e-e o seventy percent (70%) of all supports are spaced at intervals of S feet or less (note 10 feet was used in the testing program). o 95% of all supports have more bracing than the unbraced test configurations. 3.g Cable Trar Support Ouellfication Bach support will be qualified and documented by one of the following methods. totalled descriptions of these methods are presented in sections 5.0 and 6.0. 3.2.1 Testina Critical elements of supports are enveloped by tested conditions. Critical parameters o ARS o Configuration o Hardware o Cable loading Examples are shown in Figures 3-13 and 3-14. 3.2.2 Combination Test and_ Analysts To evaluate configurations similar, but not identical, to the test configurations, supplementary analysis is performed. Analysis is directed toward key support items that ensure integrity. Examples of such configurations are shown in Figures 3-15 and 3-16, 1 - -

. ~.. - -. 3.2.3 Unieue Aneirels Any support not qualified by test or combination test and analysis will require a unique analysis. This analysis ranges from a complete analysis of the support of concern to en analysis which addresses any minor deviations between the support and a previously qualified configuration. Analytical refinements, derived as' a result of the test program, are incorporated where possible, for example: 1. Experimentally determined connection spring rates are used. 2. The analytical acceptance criteria is expanded to incorporate the connection rotational and capacity limits provided by the testing program. 3. Damping values up to a maximum of 20% will be utilized. (Note: The justification to use increased damping at Seabrook was submitted to the NHC in Reference 1. The Seabrook testing program serves to confirm the applicability of these damping values.) Analytical criteria is presented in detail in Section 5.0 of this report. Each support qualification will be documented in accordance with the guidelines and checklists contained in section 6.0 of this report. 3.3 Other Issues The seismic qualification program will also address the following open NRC issues: 7

1. 10CFR50.55(e) Report Electrical Cable Tray Support strut-nuts hardware. The test configurations included bolting hardware which exhibit slippage capacities less than published allowed loads. 2. All cable tray supports at seabrook have been as-built to the level of detail shown on Figure 3-13. The personnel conducting the as-built are qualified in accordance with applicable ANSI standards. 3.4 past perfotiaance of Cable Trays in Seismic Events To further document the inherent seismic capacity of cable tray supports, seabrook cable tray support details are compared to cable tray supports that have experienced strong motion earthquakes. This comparison is made by EQE, Inc., an engineering consulting fitle that deals primarily with the historical performance of various equipment that have survived past earthquakes. EQE, Inc., has accumulated an extensive data base that includes Cable Tray Systems similar to seabrook including tray without axial bracing. Historically, Cable Tray Systems have performed very well during and after major earthquakes, with ground motion in excess of the Seabrook safe shutdown earthquake. This data supports the cone.lusions of the Seabrook dynamic test program and is provided in Appendix A. i ( l l i

~ ee CABLE TRAY SUPPORT QUALIFICATION PROGRAM IMPLEMENTATION i IDENTIFY SPECIFIC INSTALLATION I

f. li FROM AS-MJILT DRAWINOS I

t ASSESSMENT OF SUPPORT l 70 TRUCTURAL ACCEPTANCE CRITERI i I l I TESTED CONFIOURATION PARAMETRIC ANALY.. l UNIoUE ANALYSIS CONFIGURATION i I'

I ASSESSMENT OF SYSTEN t

COMPONENT INTERACTION l CRITERIA IMPAC1 WITH INDUCED LOAD ON ADJACENT CONF-ONENT CA9LE INTEORITY COUPLED COMPONENT l t 6 COMPLETE FINAL 0 QUALIFICATION I DOCUMENTATION h TABLE 3-1

l ., w i M L4J L.J. .~ l t ,vy I c-CJ e tv 8 ExtSTING d STEElenEr g 5 0W 2 ,, g v,, 'h b{ T V,1-Ta

c e.. y, g,,

r. f._ f t... s t s. v... s. 3fl LlV A~ t T y'I> a

j I
ver* REST EICTED ZONE l

i (T yp j (tv R soin 5i0E5) l h (g Tral M /g l I Tkay 9 .4 ' A '# U.h LOD Type' 1 a u aTYPe l SEE 7. s l 2 NOTE 3 6 r-9 09 SH.3C ,,/s., N i u.Wc. ;,; g - s_ .s c o, IfC."

t) top.

V ] i h 71 2 SEE N37C3 & F-9 l\\1 g .b.. -. : ~ ,.N FTYPa l %s# 8"I'A"T g l l ITYPi TRAY 1

TYPs f

fumi Sotto BTM. M I .. / L, .l3 4 t so' #~ .l TO et USED AS (

.~~

., A, fr-i

SEE woTE a.

Swius AS oE O k I P w iM CABLE TRAY SUPPORT SUSPENDED FROM CEILING [ HORIZONTAL BRACING ONLY

t 4 l.h NOTES:

..:s i:,- 47 u A bW. id! 5 58" 641".' T

1. NovEin C: ToAv5 *ER S.s::

t Costu:< ATcu :ot Tues su:"Ni es DA Y D 04 w.Lusua" r g " [ *, *s 1 L *; '. I'. .. C E L' 0 T R A n'.b assD A 4b a t tr @w4 Os2 DQ A sW'*Co S MJJ te D f._ M A uf 86.#Us.A 'Q"AaJC.E DE'TWE E4 VE R IC AL C"Jeoon j,Q' sAE ug,sRf rS uct 70 exgEp St.*. S c_ A =_a m O O K ll l. .-.SJCLD DO.W.d c.4Au D.. .. b M< A.L *.sc-Tal sem 0: TW.AY WZT*>..- 8 7 g, C-d' OE m *> JEE7b ~47 3E E*Cb.T f 7 2I rf $ As.' pie s t CO2 A2> sCA StE Of ",

17 RAsATE A.T"f ACJd M E 6JT p.

ME AR SAW m AC [ 5-COC ecAC'*JG cs2'sttusuT3 Auo rcr.se.Adcas, at!*E C 70 c,eweste watr A7.

t. -

l e .I'- W C ABLE TR AY SYSTEws ,i

Pd FIGURE 3 1 NOTES AND TYPIC A4. DET AILS l[l 4t'

'# l'f ' r3 Pre.-,.o & >.o- - :. ,w si< ~ ~ ~ ' " * '::. :.=;. ".~,'.~ - * * ~ ~ ' " o eoc m.s As a ,. up ,53l# 6 gee enemi. u.w.mc ...w r. 7 Iwa-re as a;::co ge n:W ~ 8, 8 '* 6'i e '....,.. n... .. :. :..;. : w l--l<;. 9763 t/.3DO229 SHEET I4

W3 e-6.- .aJ 3 :'. e i e.$ 1 e*.?;

f IIEE. EEI ~

t 'N g e v 'v e s* ) g g -- c.

E i

/'(*~S T/F'- (*

[ l [l O u's d' e r s me rt. 1/% W.G w.e s Os.P y e ,q SL10t NG FI T eTYPs tFei,. =E sicicTEe zone va-1 ( t v c 1:C, T ** Site *) l l I it vp) g .TYP. b 's ovi ar 1 RCCC;vto O R

1 C'.

c.... ' { stros e... N (TYP) M A c3 g o o g iType u S T^TI ~# is' sEE NDTE 1 1 1 h d' l v 8 lt \\ .gl;6. I e., \\ l i 'o o-n .,n

ustree)..

1 ns**. a t !\\ u

f. e c e.p ag J e5 LO. A~13 A A 8"> *J~ 'A E E I

no T:rusvETsk secoc:~ usu OEes e ss *> cus. -e =, :.2 ua .mo i oT.::EwEt j+ h ',, oc Tw E E 4 v7.CTid.L DO 79.N T u E vef 26. r TO T2adb ER. WOE 9ES*JT4 L.DI-D-lP try# 7 K..1 s t. slaus z.:oc war, pet so==sti uss van j J.-m. r--es / re r-m as 11 9.c c.on,::.iais cc:-u s . 1, i. 30:7.:.s: .3 c sco a umovuv es.eoEZ e t. - sr/ I ac ruvo suo ac=.a+ ao s. ouw ou psofi. : C e. -. = -7 \\ . 2, - s.Aa c ouvu De> Tao t c*~a.EE4 WCNAL. A LEDl'E evoco.~~vsvs1es g : >saoi w n ( *~w~:l - - - V -l w ncEco ao-f,s 4'" i i EwBEDDED OR SuarACE MOUNTED , J. $E E odEETs "ra?.V.PCL.TtC7 5f 2.ES I 78115df. 8.a S E PL t.T E 04 STRJ TU R STEEL l e pouca s E scacE Ago a:.-Eeva.TE - SEE NOTE 3 in -5EE SWEE T 10. ~-- an r raexaw v.s ..i rac e.ca,.,J.J:. dtE.05.!4ttut#lT$ 4.D TS.E2adcED, ~. is:.r usec so ca.seeAt unit

T.

w ..s......- CABLE TRAY SUPPORT FROM FLOOR TO CEILING HORIZONTAL BRACING ONLY WEadTarETY RE L ATED g

i. p W castr Taar sysTEus-sey p jg sof ts uC TYP8 CAL DDAILS v7 e FIGURE 3. ~'

%.gl a-(3f u*'d "0'*n o n m n.s o n as ev s:a r - W a e,5 m erv es r,. w w i s,s 'M }i t ! TIT 9753 4.'3OO22c SHEET 5 i et2 n > c..: c:... tc. c:,. m c.. 8

/ ..-.t. c Q n- = 1 lF-e.tc..::: o z -, .a ,,,. 4 .l d ~' J * '-"****' 8 - -l b ro t.'

I>dLE~, ostiz "8-1 u s.

g, st ' g* fa

n..

attacuc:2: e.a:Es tmi C /s ' < = ' , I n.t. IX (' N '~'v~'N ^ ' , "j 'r, % :r T'i F.:' I

A55 '"*
8 L y""#

<* / "" '!,"L""" ' '.h:....' '. t = - >i : E.:_.-- .a.an c <,,1 ro u 2 o r .or To oc vsto rm oc~cw. " ~- s 1 ,,Oc " '5' 86 rf 2 ^**'O C W &CAs'E FvCASE BA M ''*'Ddt?'5_EE.0"L TYPE 2 TYPE 3-T TYPE 1 o fall ATTacmastwT Ott aa tio.rAtV.<:=aiggesage,,, } 1 ~ s> u ns e' t W.c aa.a t.' LIG vttt C ' I'WIE9?-71 / \\ q JLf}-][2j{3}._JLf1 \\ 'crr. 2 .x.. T47s cc f 47 i ("U l '~4. ~ 8 8d8 '*"c "g/. '.,l 3t> agster r o og'p 2t>.f ( F ri'1 l*i W gma Q T Y P. < KA u m-m 1 -i 'l '.-) 'f g u 'u .- n rog Lapotg vetav WE est Te7C CCt ji l c m. i _ i T.'. to *G f ' L t. 1 -i s r l Sge 3 EG e-i s [ 12* EE I L m / ~ d ' SECT. *A. A' 9 g ) y a L x-lh Q S t t woT E % D w' s O %u / y-f, I ,.e 1 \\ i l lI lt l 1 ( l _,4 _a n j w l t:v i ._Z W I., t. o .r.-------g y l .x r o .,i :- e, r p i..-.. a y N i set ua U," e S EC TiO N *$ 5

5tt neoTg :

3,. u g r-4 Q ~r*> NOTES: \\. n.e a' 1 t' .*. 9 I g c,Ja.'Ex spunw= st=$F nluudTD o.t Y o= add Tat' lh s4A//t(f Or ett r'4 TFJf5 Df(/O 4. 7L7f ME'h:44 N1 *.s blLdMf fuxC,,9: al4f M dGtf f fW4'. !Mt.**% $I' lC.Y, ECCl V ETT C)

1. WO412 cm. ta t est aa t t ts S t.a t t o t 19 ( T eo affl.{9 E. & C

" ""'"'"c*s s*T t s"ip t c' c t.ats '.'.".ma. e p yjm .dnre a.<J esu s<cev ams -_- - - - --d t 'C f '* .INC-10TWl0PP ..l% 34 4 8 it a vo.D.J10 R e t t t.3 5 '* 9 T.'A'J A4f71* tac *4tPs AYTD At P #<"N D P & M -t i I I ~ ~ ~ 3."- ~.....---, "s-. " mc'= -~,"'""a,,,y, ayaye, ao,ejs7 m a a 'Q" D ~"'"' 1 9 F. o;- e.-. ". '. ', -. - - w , m t s ser s64r.s fer sem.'.7 soima.7 wr.v s E A E R O O 6< "'T* S% i - ( s r.:. r.o_ w .. ! * "= '"'m" ".aa = = re a r' a i,,,m,, %,,,",,'.",,,",", 'rm',,,., --2.m m ....mswii. o ,m, .$-ibcrni,%...,,, SUPPORT m tan an oe.c r.: vean to u . C..O.. E...,.'"d *."..U..'.'..'.."','.S, l 8 4 f WfTH 90' TURN (TYPE I ) s NyctEAR SestTY PctAT;O 6... -. >. -. -. i.,c .tu t::x '.. I,d C ABLE T8t AY SYS T E.*dG C ir%'#

'..' )

NOTES AND TYPIC At. D(TAlt.S g gfg b=lM N#W* FIGURE ;.,

p,(<.-.c L a it,,

- l D - m C *n17.'.'f.*; * "* ** p t.l.::l.

. T l ';* e ic,3.w 300??9 5aE' 7 7t

( j.s m=y...p, 3 .. ~ ( _ M* *sa r w E Tea. _ f 3

w. E: E: : g,,,,

g ~ Qd-"% rsELD DetLL tea? FLar*GE% von { l ve 13 60LT N ~ l O Tv'u9 x A i Q sad o. A. l ~l 2 45 F 3AC g s' v ~ tre AT r '* ? j l swSEat j s k r 2 f f t$ l In;% ,1 mi..1 f.' f ( tf' f f l a f' l.W I 6 n 3 \\ \\ E < ~ - m ,g g a

o J

r T...Mj 5I [I MG b M g L N ya= = s e d e . A i. i prm n'ei )1 M i r i o o 1 ~ 9 w w o -, i l S

w SECTION A A i

w -S

c e

3e w r i x 1 x 5 l l LE*T o r i l ? "@; ~ l 1 2 z lM [,l F m -L J ,r i 7 ALTERN ATE ocLDDo**f **ETMoD ?- -I. ONiv FOR SOLO BOTTong Tray T l OR wwEN a RUNG OF LaDOEe TYPE EM BEDDED I STauf l ] TAav e5 LDCATED AT A SUPPORT caoS5 MEasta. @REF1 L a q So"d*["M ) ,.3 eena-p t s- 's . :W L y e >.y.e;s-{ IGM - W l t i l l. REF ( j l .. ~ SECTION 0-8 $. 5 ALTE.'N ATE IT#av C.MI TED DAEITALY 1D CONORETE INSERT @l liFl:li mar-w Se,o~s sETwEE~ mSEnT5 is ur o wwEw v5ED. acCeivco > jg u. E. a. C. I N C. = l$ W VERTICAL TRAY SUPPORT DETAll ,i E,. 1., E. d t - --WITH HORIZ EMFEDDED STRUT i l. sEAElROOK } d S TATI O N ! ! hi NuCLEse SarEf r #EL ATED CABL E TRAf SYSTENS 4 ',.,) NOTES AND TTPICAL DCTAILS w ease g ,,,,,, y g a-4 Iwy.ei gurxorgewn._ i l { e _* s e at l7 l'" l5F63 M 3OO229 SHEET T36

Control Buildina - Switchaear Room. Elevation 21'6" 6 4 r<: =::: gg mn !E: mim q bt l E :!. 1 ' ~ r+ ^ JT .e 3 l n a wu.o2 i..o oo. L g %r; A rtyg i .y. munnew m i iu aan i f .g_ e .s. v d_[- E Q~ -- ) c j - _g y ), ~ j] g-g \\\\ 2-w% - ?; *W /_'M Q g = 5 g ;- = a '~ o y ]S {. C 1 r-E ~l'l -d. l: E ,'q a +3 3 I C s !e a i i ,pr-3 4L. E I I,p j jyl 9 + = c -=u e.~m .( ..y, 1 'r ,2 6 J J

m L

J l Z -*--- o.iz _o.zz o.zz .o. i z Figure 3.5 The typical configuration of cable trays in this area is a double trapeze with up to 8 tiers. The trapezes are connected to the ceiling using the standard connection " boot" either bolted or welded to overhead structural steel wide-flange beams. On multiple-tiered trays. at least every fourth tier is braced in the transverse direction. I Bracing details include triangular gussets at major connections and clip angles at other connections. The cable trays are connected to the trapeze supports with either internal clips or "Z"-clips. Cable is routed from the tray into electrical busses through wireways and flexible corduit. l l I 1

Control Buildino - Cable Screadina Room. Elevation 50'0" ( e-4 y.- I g i ~ y t.. G.:( y 3 l -. }} j -Zj. 3 i e 1.l ( ~ ~. ~ -} N N n* t g- ..I O O ll O y 2 4 y~ n. r > g <= b L 3 I' ( l Qw .s N l l Z *--- j

o. : c o.t 0.32 o.i2 1

Figure 3.6. The typical configuration of cable trays in.this area is floor-to-ceiling with up to 10 tiers. The trays are connected to the ceiling using the standard connection " boot" either bolted or welded to overhead structural steel wide-flange beams. The floor connection is made by welding to a base plate either embedded in or anchor-bolted to the floor. On multiple-tiered trays, at least every fourth tier is braced in the transverse direction. In addition to the large amount of horizontal trays, there are several runs of vertical trays connecting the switchgear area (elevation 21'6") with the control room (elevation 75'0").

Control Buildina - A Train Electrical Tunnel and Penetration Area. Elevation O'0" N iL h L 3 Ni ow nm, s J iTP u 3 ( I n 1 ,3 p(w: ( J v m% s C ON T awe ** f l Figure 3.7 The typical cable tray configuration in the tunnel area is floor-to-ceiling, with up to 10 tiers. One side of the tray is supported using a floor-to-ceiling column. The other side of the tray is braced with struts attached to embedded steel channel in the concrete wall. The trays are connected to the ceiling using the standard connection detail either bolted or welded to overhead structural steel wide-flange beams. The floor connection is made by welding to a base plate either embedded or bolted to the floor. Cable tray loading in this area does not exceed 40 plf. Seismic gaps are evident here. The penetration area contains a high density of cable trays. The configurations of these trays are trapeze, floor-to-ceiling, and various combinations of trapeze and floor-to-ceiling supports. In addition to the cabit trays, this area contains ducts, electrical cabinets and penetration assemblies.

Control Buildino - B Train Electrical Tunnel and Penetration Area. Elevation -20'0" nt' m I J.' c x C ~ U e. I 12t I x.- 2, 5 c TF 5 F I r.. O O r ( q \\ = =

h

\\ >.III ,T i Y tD h s @d _; 1 ' n" E388 OD J ~ i 1 1 ~""

i i

l

:=

l = = l .+ +- l" i C i z< Figure 3.8 The cable trays found here are similar to those found in the A Train Elec!.rical Tunnel and Penetration Area. l t l

Enctor Containment Buildina - Elevation -26'0" r N \\ ..Jfc; / R, \\ ~ ,y \\ MP ~ n / \\ / ( 4't 'E U l $5e .:r.

.s g

~~ s \\ k r S,.- g \\ ..e t ,/ ,i).a. / S / ,/ '\\ '~ (_-_- / N ]#l i Ro-4 i

m::: s

/ l Y l / \\ I' h. i i s o-Figure 3.9 The typical cable tray configurations in-this area are double trapeze and floor-to-ceiling supports. Most of the trapeze ( supports have 3 or 4 tiers. Cable tray loading in this area does not i exceed 40 plf. The cable trays and penetration assemblies found here are similar to those found at Elevation O'0".

Primary Auxiliary Buildino - Elevation 53'0" 6 N .a 'o s r-g .?. I e: 1I 7 n G f I 'oi b b i n- ) -l 8 i l I 1 l t e ~ ~ ~ . 0.S 2 0.52 0.52 i i d Figure 3-10 In this area, cable trays are mounted around the perimeter of the building in a 3-tier trapeze configuration. The supports are welded to the overhead structural steel wide-flange beams of the i ceiling. In addition to the trapeze supports, some of the cable trayi are supported on pedestals.

g t Primary Auxiliary Buildina - Elevation 25'0" N 'y" yS . O n,., a, w_ N 2 m m "i l f e L[ y 1 '7 E i <= o '"['. u / i .IW. -) 1 6 5 sk 'I a (3___[ _ _ l 2 6 ,s a g -l y\\ i O (3----l

,=

k w <x s 5 ? n go ~ 9g t-9

S Trr E

/ P J s 4 v Pg gg {.o.se j .. 5 e - ~ ..s e - 3 8 de o Figure 3-11 The typical configurations of cable trays in this area are double trapeze and floor-to-ceiling with up to 4 tiers.

Steam and Feedwater Pine Chase (East) Elevations 2'0" and 12'0" N n 1 tT D ls C ~' t,- q_ c- "~W) I WAWd sTEAu IUNNEL ~ CONT AINMEN T BLOG. I 1 I W l O i Figure 3-12 The typical cable tray configurations in this area are trapeze, cantilevered and pedestal supports. The cantilevered cable I trays are two tiered, with each tier braced to embedded steel channel in the concrete wall. The pedestal supports are also braced to the wall at each tier. l I l

~. FIGURE 3-13 \\ Av ange 3 d Oc /* u s g } A VJ4 ' Nf s W !#' g 21" L 12"_Lg2" ~ = \\ i k WJ6A kc.s n.23o 6 ectiq' a'~ \\ p I _ _._ _ ' I l_I i , g ' Y ~7 aggg) e:: = c = .honig'g' g s 32ff C 3' - e ce.) -sur l'### Tle el l n 9 u sy& g , p DS) q , g L en [" "* # t a k g . sg to k ' (- = r a y M F-AA9/17 NA

* f GerAr.)rusfw *'#s.'h

'w l [MMN6 M7W .$ *.] maad al 's l*n 6 " _/d4Sf/H4 WOffW j l(( H 7 - -. :& :q Nw H13 & kNx * ' s 16:$9/ 3 9.e R " 'RER 18 CERTIFID 10 INSPECT % a taar sanoars m Ess-6 CaaLE TBAY SUPPotT5 nu.SsTauem-s n.n,e, - as cartrta emocm er:

g. o. ;,.y... -

\\M WW,=///gfg Q.

VtRSTiSSUE A f Y"*

i / / DATE -...,--m. ,._..m

~_ \\ FIGURE 3-14 'l i i r~ e 6-0 l 2-G 2 t-i i Tos St 2l rob P . N.. Z i 'v ( 4:~ s, -I v cs es i Det.l D-3 f ac t. S S e r 7 r 7 Tyr' 3 ft.n -;( , ~a G.. E O ra 4 r 7 r L E D C7 EL16 o r y for bu,. +l, '.*.l a < l S e e De 6 4 S ect SE-i SECTood SE-3 ~ w90 'or 211 15 CERTIF1tD to INSPECT CASLE TRAY SUPPot:5 h i 80?PatT8 M R ESG-4 - AS-CDESTRLTTED - ( ee BLDC, CW7a4 r EL. #

C M 1MtB ms M L N ST:

r: e* i f re f % gj fm&T ISSUE ced E " " *- 8 hTE ...i l

n :....,

4 s,y 97gpt.37014' 5tt.7A 2-y

FIGURE 3-15 a. N NAT'O N es. m e. Umastne=== * *, " * ' ' r pa n>a t e a [q,, g "o,g,, poe+etasseoca sTAtiose ias c

==-- e,-- m As.Le a_n (?m Am Gt 11-4

9 % 3o101 ET1*aa W3 REFERENCES PIN.9763-800012 4 D

r e ,a- ,-e n- =<s. = = + R-A wW W 14 s.. W ~ y, n.s jnu j y 3-s ~ ryrns A H M] Y , (m >. go gg r t y pj ~ 6 reacts) ^% 1$. M grp g g.g g 5'M A b of P j r,,, 7--) u n uverg l ; aw

,8 ra" r

a =<, l h 4"A y* raw o o, a w,,,, wwy m,3 u anw r n.=.,, .,1 v,. e -/ 3< n r n w,- 4 r f w&' a t2 q [] 3 ,,a ,,.,s (= g 8 %,, ' 1 7a< u l'% Ry= r, 2 we.puses L n'8 M 88 I-upAgotw sW d' Later WP& M UN,pr)ost rec D6e.rso c 3 sat n' 6 as wr W** rp Ks a = 118 6-% carineJ %)-3 ( P4fnad e, nos % ar) g s Tn'5 ~ QU Dutd: 3:04+9 i l TA: WO(,9 'm'. P. u 8 l 'ECKER IS CERTIFIED TO INSPECT CABLE TRAY SUPPORTS CABLE TRAY SUPPORTS PER ESG-4 - AS-CONSTRUCTED - , et SLDG. 004)T40L EL. El ~4" CERTIFIED AS-CONSTRUCTED BY: ,........c s c o. e e a n =a m e.ain e L /CA rz 7c o a esasinssus se w 0 ~ ~**" '-- ~~

'*'*'**"MM%=.eww-e a M4 m og e,.g FIGURE 3-16 i ~ oc, = co uririo r EleREDLWu me W ; g _jfc_ ~ e' 8 Aw W4 PS'e+stAamoon sTirioN m iaa cc> span, sus 4cr M-Dasratisa daurs.at drJM. (2 21'-6* de 1PN d e /0J. J&CW S.l-h j cam REFERENCES PIN.9763 SQ0012 O e a n ... - +.c S 2 ')t," n= r

P, 24

_ w v_. M = w u_e_._,__: s._ _a s 9' \\ pp 9 1 e N$$ % ./ (ygg ey f y "a czm ces n L, m m + "%iA/J;i?. 4%,)l M' h l _ to of,. 3344 -1 at E Rolf l w u.m (. 4 a.4- -a go n, s ,,,,,S p!p'* "' s ]; r s a ""U"A M I 8#U-6 4 ,o l of I M l {~ l l SMM f, le 5 f gggy*7 "I .u'a" l 7-agg j

, L.

n J u a.J f ggg w' 7 I dQ l [- EN#*3 33 3, bL "]. I kj64 ^4 'l N l kbf \\= Mj -ungnu N

BAE433DS (8'O ds

pun,

@c St 71P6 As TJr o at as. imew/tur,% gas) as. ue a m i S$48 bf 7M3 a.9es a.nc E~OLgNnau T.tmi WT) .,anos

- Tus

_ f.. _._.. g TA: MG9 REV.I CEAT/FIE> To ESS-4 Rea L. Ct.q wit.O o E.9. f U.P.0 6 [ N. & f-30 Q Qgg, GQ r h.y / 7.CKER IS CERTIFIED TO INSPECT CAiLE TRAY SUPPORTS wa8LE TRAY SUPPORTS PER ESG-4 - AS-CONSTRUCTED - BLDG.Co Ai74ot. EL.2s'-fo,, i 2 'k/g, SEdi'A o M. W if-c5 H4 7MM CERTITIED AS-CONSTRUCTED BY: g,y,,,3, A s N,o rt h. w uc ... pe. ...c e..,........... i m x. 2 o i A Ed4Vs2-1s a g nasT issue p & ~ =*==' -

4.0 TEST PROGRAM DEVELOPMENT 4.1 Test Program The Cable Tray System qualification program includes the development of the items identified in Section 3; specifically, dynamic testing of full scale sys, tem models, connection tests and analytical applications program. In addition, the program data base was enlarged by the utilization of an existing Bechtel Raceway Test Program. The detailed development of these programs and a summary of ensuing results is included in the following sections. 4.1.1 Dynamic Test Program The primary objective of the Seabrook specific tests was to study the seismic resistance and response of typical multi-tier Cable Tray Systems, constructed uring representative site-specific details and hardware subjected to various levels of postulated seismic loadings. Other objectives include the collection and analysis of data to detenmine trends in resonant frequencies, damping ratios, response shapes and support loads. This information is later coordinated with an analytical applications program to qualify the support system as described in Section 4.2.3. The test effort investigated the performance of the typical Cable Tray Systems using three different load sequences and input levels. They are identified as follows: (Reference 10, Vol. 1.) SSE Test Each test configuration was subjected to one SSE Test Response Spectra (TRS). To study system response, the test configuration was also t subjected to fractional SSE events. _g-

Fatiaue Test Each test configuration was subjected to five OBE TRS followed by one SSE TRS. Fraallity Test e Each test configuration was subjected to incremental TRS, using the SSE TRS shape, until the table limit was reached. This resulted in an application of spectra ranging from 1.2 to 1.5 of the SSE TRS. Each test sample was subjected to a specific sequence of test inputs to optimize data and performance evaluations. All the test cases include low level random testing, fractional SSE testing, SSE testing, fatigue testing and betting studies. Details of the test sequence for each test can be found in Tables 4.1 through 4.3. Test plans for each test are found in References 2 and 3. 4.1.2 Development of the Dynamic Test Confiaurations Site walkdowns, performed to support the Bechtel damping study of the Seabrook cable tray supports (Reference 5), were used as the basis for the proposal of the two test sample geometries (Case A and B). A description of these two cases are provided in Figures 4-1 and 4-2. These two configurations were judged to be the most representative of site conditions. Two test cases were selected initially, as it was felt that additional test cases could be added as warranted. The main Seabrook typical supports are represented by the trapeze support T26 (representative of T4, T8, T26, T27 and T29) and by the typical support T5 (floor to ceiling support). A sketch of typical supports, T26 and TS is provided in Figures 4-3 and 4-4. These two types were selected for dimensional, quantity and behavioral reasons. These initial configurations consisted of reduced transverse and no longitudinal bracing, which is consistent with the program objectives. -

~ - - ^-" ^- ---.~1 After the completion of Test Case A, it was decided to establish a third configuration, Test Case C. This test case is illustrated in Figures 4-5 and 4-6, and incorporates two major types of supports (change in direction rupports T9, T10) and supports for vertical tray, T38). The third case, by enlarging the data base, serves the joint purpose of illustrating the seismic resistance of two different distinct types of support conditions, while providing additional data to aid in the analytical applications studies. A sketch of typical support, T9, is provided in Figure 4-7. Testing was performed almost c4clusively with site supplied hardware. Due to shortages of necessary details, locally supplied hardware was used, however, only site representative vendors were utilized. All the test samples incorporated various representative in situ details which were determined to be particularly sensitive to the increased motions of unbraced systems. Table 4.4 provides a list of typical features simulated. Typical connections, anchors and tray fasteners tested are shown in Figures 4.8 a, b and c. Further, in addition to these representative in situ simulations, several items were maximized for conservatism and margin. In addition to the reduced bracing previously mentioned, the primary features were: a) the use of all l trays at a forty percent fill (by area), and b) tha use of a maximum cable tray support spacing (10 feet in lieu of typical 8 feet or less). All testing was performed using an envelope spectra. Thus, the Test Response Spectra (TRS) generated to envelope the Operating Basis Earthquake (OBE) and the Safe Shutdown Earthquake (SSE) envelope all applicable sito conditions. This is extremely conservative and it must be remembered that at certain elevations, locations and directions in the plant, the local Amplified Response Spectra (ARS) is a fraction of the plant envelope TRS. In most cases, the applicable design ARS is only one-third to one-belf of the TRS. Testing to the TRS, then, provides significant margin. (Figure 4-13) i The use of envelope conditions (maximum cable load, maximum support spacing and envelope spectra) were selected to give the testing program the most flexibility and broadest applicability possible. The program objective L 2 is to envelope the behavior of as many supports as possible. The comparison of typical conditions (vs. envelope conditions) is identified in Section 3., f .--,.y-- -_,____,.-._.,._,-__-m

4.1.3 Connection Tests In addition to full scale dynamic tests (Reference 10. Vol. 5), load deflection stiffness tests and cyclic fatigue connection tests of key representative connections have been performed as an integral part of the qualification program. The connection tests were developed to study the performance of typical cable tray support connections and to provide data to re[Ine modeling techniques, and establish appropriate connection stress and rotation limits. The connection test plan is found in Reference 4 Past Bechtel Raceway Testing (Reference 6) established that the behavior of the primary connections is a key parameter in determining the system behavior of Cable Tray Support Systems. As discussed earlier in Section 3, the data collected is used as input to the cable tray support analytical models. Specifically, spring rates are developed to model these connections in various finite element models. Three test types were performed: o Moment resistance as a function of angular rotation (M versus B). o Load resistance as a function of deflection (p versus A). The connections tested are shown in Figure 4 -9. These connections are representative of the anchor connections found in the field and simulated during the shake table testing. The test Jetup is shown in Figure 4-10. The connection test system consists of a rinid vertical surface to which sacrificial plate and strut assemblies can be attached hori=ontally (as shown) or vertically, permitting testing of each anchor detall in its principal axes. The following equipment is required: I o A double acting hydraulic cylinder. o A stiff mer.ber (6 inch x 6 inch x 1/2 inch) to limit strut deflection, o A load cell to sense load. { l 1

o A linear potentiometer to sense deflection. o A mechanical force gauge to sense applied axial load and a group of tension springs to apply that load. 4.1.4 Coordination With Bechtel Raceway Support Program e New Hampshire Yankee is also a member of the Bechtel Joint Owner's Group and therefore the data from Raceway Support Program (Reference 6) is coordinated with the Seabrook testing. Bechtel (Reference 5) previously reviewed the proposed Seabrook design and evaluated the design for applicability of the test data. Because of the generic nature of the data, it was determined that the Seabrook design was bounded, and the data was applicable. This data has been presented to the WRC and their initial concurrence received in May 1985 (Reference 7). The design damping curve for the Seabrook Project (Figure 4.11) was developed from the results of the Bechtel sponsored, " Cable Tray and Conduit Raceway Test Program" performed by ANCO Engineers, Inc. To date, in excess of 2,000 dynamic tests have been performed as part of the i,achtel Cable Tray Testing Program. Numerous tray support systems have been tested and the effects of a broad range of parameters have been investigated. Flexible as well as rigid support systems have been tested. Results of these tests have demonstrated that the Cable Tray Support System damping is greatly influenced by the amount of motion of cables in the trays. For linear dynamic structural analysis, the effects of the various mechanisms which tend to dissipate the energy of a system are typically lumped together in a single factor known es the effective viscous damping. This velocity dependent parameter is connonly quantified by means of dynamic testing, and can include the effects of many energy dissipating mechanisms, such as friction and rotation in bolted connections, hysteresis, radiation of energy away from foundations, and others. The pre. dominant energy dissipating mechanism observed during the Cable Tray Test Program was the vibration of the cables. A significant amount of 1 l L

~. _.......... energy was absorbe'd as a result of friction between adjacent moving cables and between cables and trays. An equivalent viscous damping was calculated for each tested system, based upon the recorded dynamic input and response. A detailed discussion of the damping computations can be found in the test report (Reference 6). The individual damping values clearly demonstrated that the tested cable tray supports comprise a dynamic system with high equivalent viscous damping. Results from tests of many support types and configurations are included in the data, but in the interest of providing a generic design damping curve, the conservative bound of the accumulated test data, represented as a bilinear curve, was utill::ed. Variations in support rigidity did not significantly impact the system damping, that is, damping data in excess of Figure 4-11 was still realized. The effect of the cables on damping is heightened with increased input acceleration levels. When cable trays are lightly loaded, or when the system is subjected to low input acceleration levels, the measured damping approaches the values in NRC Regulatory Guide 1.61 for bolted structures. The tested support systems were constructed using standard cold-formed struts and standard bolted fittings from a variety of manufacturers. Cable trays and fittings for the tests were provided by several manufacturers, including Metal products Corporation, which is the sole supplier of cable trays for Seabrook. Tests included trapeze supports of varying height and with various transverse and longitudinal bracing configurations and rigid supports. Cable loading ranged from 0 to 50 pounds per foot. The fundamental frequencies of the tested support configurations were found to be in two rangeJ: the more flexible support systems had fundamental frequencies of 2 to 6 cycles per second (cps); the more rigid support systems had a range of 9 to 25 CPS. The latter frequency range results primarily from tray variations, as the supports the.selves were comparatively rigid.

The wide variety of the tray types and support configurations included in the test program simulated actual field installed conditions. A large number of variables were investigated, including: o Types and manufacturers of trays o Type and size of tray supports o Location of tray splices o Number of tray tiers o configuration of support systeam o Type and spacing of transverse and longitudinal bracing o Weight of cables o Cable ties Extensive dynamic testing of the effects of these and other variables has produced voluminous raw data, which has been summarized in the test report. In view of the scope of the test pectram, it has been concluded that the tests simulate actual field conditions, and that the results are applicable to the design of comparable tray support systems. The testing program clearly demonstrated that a significant portion of the support system damping was a product of cable motion and the resulting friction between cables and between cables and trays. Therefore, in order to assess the compatibility of the Seabrook system and the tested system, the frequency and general characteristics of the Seabrook system have been studied to determine whether they fall within the bounds of the test program, thereby providing assurance that the cable motion necessary to produce the predicted damping will occur.

As a result of the Bechtel Application Study, the generic results of Raceway Test Program have been applied to the Seabrook Station. The more important conc!usions are tabulated below: a. Cable Tray Raceway Systems have damping that rantes from 15 to 50 percent for trays with cable loading from 70 to 50 f/ft. Below 20 #/ft, a reduction in damping is observed. The lowest damping was for unloaded trays which have damping more closely approximating the 7 percent permitted for bolted steel structures by USNRC Regulatory Guide 1.61. Input motions (both vertical and horizontal) excite cables within the loaded trays and cause them to move relative to the tray. This movement is~either a bouncing or sliding of the cable within the tray. The motion of the cables appears to be one of the energy absorbing mechanisms that contributes significantly to the high damping valves, b. Damping tends to increase with increasing input. At response levels anticipated during strong Earthquakes (0.2g and greater SSE), raceways are so highly damped that they respond to a broad band energy input rather than to a narrow frequency band energy characteristic of a resonance condition. c. Anchor point flexibility, as determined by the connection details, can be more important than the flexural stiffness of the struts in determining lateral frequency of systems. This indicates that the tested connection details result in a partially-fixed condition at the anchorages, d. Typically, cables do not appear to influence overall system stiffness and consequently only the mass of the cables need be considered in computing system dynamic responses. The recent shake table testing of the Seabrook support systems reinforces the inclusion of the Seabrook system. Using site supplied hardware and site-specific configurations, similar system conclusions and damping data (see Section 4.2) were obtained. Therefore, the entire Bechtel data base, in ~ , - ~- l addition to the Seabrook test data can be applied to system evaluations at Seabrook. The actual damping values to be used at Seabrook as presented to theWRCthispastMay(Figure 4pl)remainsthesame. These conclusions are confitined by ANCO. See Reference 10. Vol. 1. 4.2 Program Results e 4.2.1 Dynamic Test Results The seismic simulation tests (Cases A, B and C) have been completed. One item to note prior to any discussion of the results is a discussion of the Test Response Spectra (TRS) and it's relative magnitude in comparison to the Required Response Spectra (RRS). The RRS envelops the floor response spectrum of the seven seismic Category I buildings within the scope of this qualification program. Figure 4-12 represents a typical TRS versus RRS comparison plot. The TRS is very broad and contains greater energy than the RRS. In addition, although not a complete site envelope spectra, the TRS possesses significant margin when compared to specific site locations. For example, Figure 4-13 depicts an SSE TRS versus the control building floor response spectra (E-W, elevation 21'6" to 50'). The test enveloping is very conservative. The above building floor response spectra may also be subdivided into area local ARS which further increases the conservatism of the test envelope. i The primary results of the seismic simulation tests is as follows: 1. Seabrook specific testing results were in agreement with.the i results of the Bechtel Raceway Support program (Reference 6). ? 2. With increasing levels of input, some minor damage was observed. Table 4.5 illustrates representative damage as a function of percent SSE input for the "non fatigue" earthquake testing. Recall, however, the conservativeness of the SSE envelope RRS and the TRS enveloping. i 3. All primary connections survived all testing, thus ensuring overall system integrity. i

., !L'. - ~ l 4. Th'e measured damping for all the earthquakejte. sting was in excess of 20% critical. Tables 4.6 and 4.7 provide',a summary of a portion ~ of the available data. 5. The tested cable did not exhibit any physical wear or damage. No losa of continuity was observed when mot.itored. e 6. Overall integrity was demonstrated for the three test samples in both braced and unbraced configurations. 7. Table 4.6 provides a sumary of the maximum support displacements (relative to the test table) due to "SSE" testing for Cases A and B. Case C, although not shown, was less than case A or B. For illustrative purposes, representative displacements are also provided for Case A. These are displacements due to the TRS input shown in Eigure 4-13. 8. Horizontal brace loads were more effective in resisting seismic loads than the diagonal braces. Table '4.9 illustrates.the degradation of the braces during fatigue testing. The degradation stems primarily from the geometry of the connections for diagonal braces and the fact they are two-bolt clips. It should be noted that (1) diagonal braces were effective when used with horizontal braces (Cases B and C) and (2) the TRS conservatively enveloped the SSE envelope RRS. 9. Fatigue testing (at the envelope OBE') did soften the mystem and effect load distributioti. The qualification program will consider the load distribution observed from the testing pror, ram. t l l

10. Maximum stresses (peak) recorded in both the brace and post during each test were typically less than 5 KSI, which is well below allowable stress values.

(See Table 4-12.) l l \\ ! l l p,

4.2.2 Connection Test Results As introduced in Section 4.1.3, representative connections were tested to obtain rotational and translational stiffness data, and cyclic fatigue data. The data was required to a) assist in the development of finite element models of the shake table test samples and the Seabrook installations and to b) to establish performance requirements. A sample moment-rotation curve for the 4-bolt gusseted angles (Detail 33DU, Figure 4-8a) is provided in Figure 4-14. This data is typical of the connection tests in that it demonstrates: (a) ductile behavior and (b) a significant connection moment capacity. These results are very similar tc the results of various past connection tests performed by Bechtel (Reference 6). The fatigue tests document the large number of cycles these connections can undergo. This, also, was demonstrated in earlier Bechtel testing. The Seabrook data enlarges the data base and results in conservative predictions of connection fatigue performance for Seabrook. The typical shape of a fatigue plot is shown in Figure 4-17 The number of cycles to failure (N) is depicted as a function of a constantly applied rotation (G). 4.2.3 Correlation Analysis of Test Samples 4.2.3.1 Analysis of Test Samples The analytical modeling of the four test cases is an integral step in the qualification of the in situ cable tray supports at the Seabrook Station. i The shake table test data will be combined with the connection test data and used to demonstrate that the refined analytical models will enable realistic L system evaluation. l l I l As discussed in Section 4.2.1, the shake table test program demonstrated two distinct features associated with the cable Tray System j dynamics which must be addressed: joint flexibility and amplitude dependent frictional losses due to cable vibration, etc. Analytically, the treatment of l l f 1 l

i amplitude dependent frictional losses will be approximated by the use of realistic damping values and joint flexibility based on moment-rotation test data. Linear finite element models have been developed for test cases A, B and C. Preliminary spring rates have been developed for the connections and have been inserted into the finite element models to simulate joint behavior. e Tables 4.10 and 4.11 demonstrate the ability of the analytical models to closely correlate with test results. All the models represented in these two tables utilized identical spring rates and damping, therefore, demonstrating the consistency of the techniques involved. Other than incorporating these two effects in the finite element models, the modeling techniques are unchanged with relation to existing project techniques. Preliminary analysis has demonstrated that the use of the previously described springs and damping valves result in a system which correlates well with the test results. The analytical results indicate that Cable Tray Support Systems are effective because they resist lateral loads by framing action. The correlation of the test and analysis illustrates the effectiveness T of the refined modeling techniques to predict the dynamic properties of these representative models. The correlation of analysis and test results for the test sample effectively serves as a program link which combines the connection tests and shake table tests. The correlation demonstrates the importance of the connection behavior, validates the refined analysis modeling and provides the data required to establish a connection performance criteria. 4.2.3.2 Application of Connection Test program The Seabrook dynamic tests (Section 4.2.1) clearly demonstrated the inherent seismic capacity and energy absorption capabilities of the tested Cable Tray Support Systems. As stated earlier, these three samples were ( selected for dimensional, behavioral and quantity reasons. All the system t l geometries that are installed in the plant, however, may not be bounded by the 1 three tested samples. This necessitates an analytical program, wherein the support systems not already geometrically bounded can be evaluated. 1 _

The shake table tests presented a bounding condition within the limits of the test facility. The tested geometries were subjected to seismic inputs in excess of plant envelope conditions, braced and unbraced conditions, maximum cable loadings, maximum support spans, maximum support widths, hardware variations, etc. These severe conditions and the ensuing performance are indicative of the available margins exhibited by the test samples. It is fe(tthatthetestsamplesweresubjectedtoresponselevels(stresses, strains, displacements, etc.), which will envelope the large majority of site conditions. Thus, the primary purpose of the analytical phase of the qualification program is to evaluate the plant conditions to assure that the acceptable response levels demonstrated experimentally will not be exceeded. As discussed in Section 3, the existing plant evaluation methodology must be slightly amended to accomplish this task. The details of the amendments necessary to evaluate the connections is described in the following paragraphs. As the existing project criteria does not address certain joint flexibilities, a criteria is added to document the flexibility of the key connections of the support systems. The use of connection springs in raceway-modeling has been implemented previously by Bechtel in conjunction with the Bechtel Raceway Test program (Reference 6) and has been reviewed and accepted by the NRC staff. The NRC staff has considered Bechtel submittals of design guides, sample calculations, sample computer models, etc., in their past reviews. References 8 and 9 document past NRC programmatical acceptance in the form of plant Safety Evaluation Reports (SER). Because the connection springs represent a new feature in the Seabrook analysis program, an acceptance criteria consistent with the existing criteria is required. The nature of the connection's behavior and cyclic loading requires an evaluation of not only the allowable load and displacement, but also an evaluation of the connection's fatigue capabilities. This is, again, in conformance with past methodologies utilized by the Bechtel power Corporation. Detailed methodologies have been submitted on several occasions; and therefore, only a brief summary is provided here. A sample behavior of a typical cable tray support connection is illustrated in Figure 4-14. As presented, there is typically very ductile behavior and substantial reserve capacity. The criteria proposed for the. --

l 1 analytical program would conservatively neglect these attributes since connection rotations are limited. The connection spring stiffneas, , is a conservative application, as it underestimates joint strain energy capacity and the response is conservative until the rotational acceptance criteria is exceeded. 4.2.4 Analytical Configurations Analytical models developed to evaluate the cable tray supports at the Seabrook Station will utilize the criteria identified in Sections 3 and 4.2.3.2. Because of the similarity reflected in many of the supports, parametric models are generated. As discussed earlier, the qualification program for individual supports entails a review using test and parametric analysis results. Supports not qualified in this fashion will utilize individual evaluation. The analytical models discussed above have been generated fram the results of a site walkdown and review of as-built drawings. The site walkdown and as-built review were performed by a team of engineers f.om YNSD and Bechtel, who were familiar with both the testing program and the Seabrook cable tray layout. The purpose of the walkdown and as-built review was to identify support configurations which were not geometrically enveloped by the tested configurations. The following parameters were considered: o Width of the support. o Support spacing, o cable fill. o Type of connection used at the building attachment. o Other considerations (custom configuration, miscellaneous attachments and close proximity items)...

As a result of the as-built review, several configurations were identified for parametric analysis. These configurations are representative of the support types, size and loading in the buildings under consideration. 'Each of these configurations will be analyzed using data obtained from the test to determine system response and load path distributions to primary support and bracing members. Figures 4-15 and 4-16 depicts typical well-mounted configurations to be modeled for the analytical program. I I -

TABLE 4-1 TEST SEOUENCE - CASE A o Preliminary testing, low level random input to determine frequencies, mode shapes and damping ratios o Earthquake testing of configuration A at (3) fractional SSE input levels and (1) full level SSE. o Fatigue testing of configuration A five OBE level events followed by a single SSE level event o Remove all bracing from test configuration. o Preliminary testing, low level random input to determine frequencies, mode shapes and damping ratios o Earthquake testing at SSE level and at Itvele eaceeding SSE level. 0 e O

t LLEE 4-2 TEST SEQUENCE - CASE B o Preliminary testing. Iow level randon input to determine frequencies, mode shapes and damping ratios o Earthquake testing of configuration B at , 2) fractional SSE levels and (1) full ( level SSE. o Fatigue testing of configuration B; five OBE level events followed by a single CSE level event o Earthquake testing of configuration B at a level exceeding the SSE level o Remove all bracing from test configuration. Preliminary testing. Iow level random input to determine frequencies, mode shapes and damping ratios. o Earthquake testing of unbraced configuration at SSE level. o Disconnect bottom connections at ' floor to floor' supports. Preliminary testing. Iow level random input to determine frequencies, mode shapes, and damping

ratios, o

Earthquake testing of unbraced configuration at SSE level

TABLE 4-3 TEST SEOUENCE - CASE C o Preliminary testing. Iow level randon input to determine frequencies, mode shapes and damping ratios o Earthquake testing of configuration C at (2) fractional SSE levels and (1) full

1evel SSE Fatigue testing of configuration C: five o

OBE level events followed by a single SSE level event o Earthquake testing of configuration C at a level exceeding the SSE level o Remove all bracing with strut connection hardware. Install longitudinal braces between 53 & S4, Install transverse brace at 53 with welded connections. Earthquaking testing at SSE level, o Remove all bracing. Earthquake testing at SSE level.

~ TABLE 4-4 SIMULATION OF VARIOUS IN-SITU CONDITIONS D + o Site supplied materials (strut. hardware. cable ties, etc.) o Typical wireway attachments o Alternate tray hold down hardware o Mixing of tray hold down hardware o Eccentric connections o Vendor variation of connection fittings o Standard primary connections o Tray type variation o Cable ties o Cable tray voltage levels o Vendor variation of strut nuts Typical cable tray spli'ce and cover details o Variation of horizontal brace connection o details Cable tray elevation transitions o Cable tray direction transitions o Conduit to cable tray interface details o Forty percent cable tray fill (by area) o o Bolt torque values

s TABLE 4-5 OBSERVED PERFORMANCE (REPRESENTATIVE) 1 Approximately 40% SSE Input Level o No visible deformation of connection o hardware Isolated bolts I.,s s some torque o Appioximately 60 to 70% SSE input level 0 No visible deformation of connection o hardware o A few bolts lo,s e some torque o Isolated bolts (typically diagonal braces) become loose Approximately 90% SSE input level o o Isolate'd 'Z-clip' deformation o Slightly more bolt loosening than previous category (a few bolts become loose) Approximately SSE input level o o Cable tie breakage o Multiple 'Z-clips" deform Acute brace clip angle deformation or o fracture 1" g conduit clamp movement o o Tray slippage (w/ mix of bent and*Z' clip usage) Approximately 100% to 130s SSE input level o l o Many cable ties break o Overhead gusseted angle weld fracture o Overhead P1000 deformation Vertical slippage of horizontal member o Minor horizontal member rotation o Isolated internal connection visibly o l deformed l o Diagonal brace and overhead connection bolts lose some torque 0 l

\\ D TABLE 4-6 DAMPING - CASE A TEST LONGITUDINAL TRANSVERSE 7.3.1.2.6 - PARTIAL SSE 22% 36% o W/ BRACING 7.3.3 OBE W/ BRACING 22% 25% 7.3.7 SSE AFTER (5) 23t 27% OBE W/ BRACING e e

=&'9MM = s m. m g,%,,,,. e o G D TABLE 4-7 DAMPING - CASE B TEST LONGITUDINAL TRANSVERSE 7.6,0.1 - PARTIAL SSE 30% 27% W/ BRACING 7.6.2 OBE W/ BRACING 32% 28% 7.6.7 SSE AFTER (5) 33t 32% OBE W/ BRACING i I O ( e l l

j e TABLE 4-8 MAXIMUM DISPLACEMENTS BRACED BRACED UNBRACED (FATIGUE) LONGITUDINAL 4.3" 5.2" 4.6" CASE A TRANSVERSE 2.6" 5.2" 3.8" LONGITUDINAL 1.6" 4.0" 2.0" CASE B TRANSVERSE 2.7" 4.6" 3.8" l l REPRESENTATIVE DISPLACEMENTS CASE-A (BRACED) TEST - 7.3.1.2.7 ( 70% SSE) LONGITUDINAL TRANSVERSE l 1.92" 0.96" l l

I~ ~ '1 8 e D TABLE 4-9 HORIZONTAL BRACE VS. DIAGONAL BRACES SUPPORT #1 CASE A - DIAGONAL BRACE CASE B - HORIZONTAL BRACE CASE B CASE A TEST 4000H 3800H OBE M1 3300H 3000H OBE M2 3000H 1000H OBE M3 3000H 500H OBE M4 3000H 500H l OBE M5 4500H 750H I SSE l 1 ~ ~ ~ '-~ '

_ _.. ~. TABLE 4-10 FUNDAMENTAL MODE C0pfARISONS (w/o Bracing) TEST DIRECTION FREQUENCY ANALYSIS TEST CASE A Longitudinal 2.0 1.8 Transverse 2.5' 2.6 CASE B Longitudinal 3.4 2.9 Transverse 4.2 4.3 ~ TABLE 4-11 FUNDAMENTAL H0DE COMPARISONS (w/ Bracing) TEST DIRECTION FREQUENCY ANALYSIS TEST CASE A Longitudinal 3.6 3.3 Transverse 5.4 5.7 CASE B 'engitudinal 3.5 3.6 Transverse 5.2 6.0 CASE C Lon>:itudinal 5.1 5.8 Transverse 5.1 5.6

MAXIMUM TEST S1RESSES SSE (with bracing) CASE A BRACE 3.3 (KSI) POST (TOP) 2.7 CASE B BRACE 3.6 (KSI) POST (TOP) 3.9 ALIDWABLE STRESS 0.9Sy 40 KSI = = TABLE 4-12

i i 2 \\ i I t I { gf gf gf'* m'n'* i ay i i __ ___ _ __ _ mme i _._ _ _ _ I muurt M mW' mamm-W F= _ .i 4 N.,%W N W =W w .~G,% W Y i \\ l l i FIGURE 4-1 TEST CONFIGURATION (CASE A) l 1 l t r e t

~__-.__. ._~.-..-___. 4 Y I! 'df .9

= =

YW k m%ma ~ %F ,+ sw r 1 FIGURE 4-2 TEST CONFIGURATION (CASE B) \\ r I i i

SEABROOK CABLE TRAY SUPPORT TYPES CONC OR STEEL ~~- /////////////////> //// ///P CABLE TRAY (TYP) 30* TRANSVERSE (TYP) BRACE n M 5 o P1004 pCABLE TRAY ri (typ) ( TY R) a A "\\ TR AY SUPPORT P1001 W', l "L" (typ) (WR) FRONT ELEV SIDE ELEV ~ ) N { ,/ ' ~ / s s / ~ \\, TRANSVERSE BRACE / / / / ~ NOTE : "L " VARIES FROM 5'-O' TO 10'0" BRACE TRAY) FIGURE 4-3 TYPICAL TRAPEZE-TYPE TRAY SUPPORT TYPE @

SEABROOK CABLE TRAY SUPPORT TYPES ~ isn. /// m 'n' } t TRANSVERSE BRACE CABLE [/ i ~ TRAY 1_* i: i ? x2 n m P1001 '/ e j i (Typ) 7 A_ L t-- i. CABLE M P100 f. i S_ _ (TYP) f Q; TRAY j (TYR) / u g j 7;;~ ~ ~ 7er - ~~ / /,< < is./ _W 10*-0" _ / I; "L" l z' (MAX.) f FRONT ELEV SIDE ELEV / / 2^ """ S" /- 9,/h/ A NOTE: "L

VARIES FROM '5'-0" TO 10*-0*:

(TYP RAY - / /- % j/' k FIGtfRE 4-4 TYPICAL FLOOR TO CEILING TYPE TRAY StJPPORT TYPE @

9 i i 9 3 + h a_ i b i te g-s i e I d j ,g E _4 1 ~ y l 2 l l r ._l .l \\ W i

~ 4 9 5 L _. en. *,.W-, sa k_ Q, N .I -a m l mem n~- "5 W m m-m 2 r a s: ~ ~ ~ ~w l l mj J I - P FIGURE 4-6 TEST CONr!GUPATION (CASE C') e 6

~ SEABROOK CABLE TRAY SUPPORT TYPES ~ ~ b x.: - y .n:s y 1 e i .v- ,~ ,L _~ ^ w-1 " r"' 9 3 N O \\ k0 i=5 ,9 p<cm. rw n%u (TYP) 'x n c ~ ~ a p, -^ ~^ , w = 7 i=1 pix 4 - (YYP)

=d

?r .2 A a:o w ex ~ SiOE El.EVA TICAJ TYPICAL TRANSITIONAL TYPE TRAY SUPPOPT TYPE @ FIGURE 4 7

e, x i '[z 3_' &Q~T i'f$ ~~T~~"J'$# ~~ 1 ~ 0 0 n l ll -40 04 9 4 N~1Y: 3 33FO \\\\ l 1 33fD I \\ i:" '3300,7YP 'y l t s l l ll 8 i l i ii1 .!I ll 8 l l 3I l !III I I II 1 II l I11 !j! d _Lijl_!_ _t_L,LI. i i AUCHORS i rTin- , i i; 9 i r i-8 8 IIi ill 3 l1:

i, iii i iii i

I 1 S 11 i i Y ji ll -k'tii ~ ~~ l.! T'il, I I 33oqM; r Ii --[i 33o:-- t 1Ii _ 4,I 3 i I i 35cv at t! I i pi i INTERMAL COMUECTIOA/S TRAY SUPPORT CONNECTIONS FIGUFF 4-Ra

i. 'l ~ 2 I j llI l f 35Df/,TTP

i n I: _

l g lg 4 1

~_

j it "33 DC.TYP , b - -g - - -- - - - - -a,- -f in l3 h ltIl i ll li, -11 \\ \\ i I l j s s s. 33eo is i -- i t \\ [i i, 33EA gs \\ I l l 4 g\\ \\\\ li Ii._ w \\ lI l.1_:t i 1 4.T ~ j \\ g 's \\ \\ ll it 1I -t-- l \\g \\ li 'F- .I ! : 1i 4 I' BRACE COMMECTIONS_ ] l'l.I: sau,rrp 1 ] l'l l7 I I i +l r 3sonryp ii1 _f.__J lii 1II g Illl n____- 11 i a ll 1 ii g_w___ II 4_ .tf jl II 4_- l 4-- I! _ t_Lp 1: f, ~ BRACE CONMEC770HS-t TRAY SUPPORT CONNECTIONS FIGURE 4-8b i

i, i I r-I I I I 8 M i r r-33CL N/ 928 92C 3306 3SCM TRAY FASTENING HARDWARE l FIGURE 4-8c 1 I l I 1 i i

'$~

1 Ii iI lI Il 1

33 M i I I Mvs. G; Pvs. h g = = /// ///////////////// II I l l I tl l 33 FD ii 1 s Mys. G; n/'.wo 90 P/Als 4.0 0.L ///////////// /////// ..g \\ 33 04 l l lf ll l M vs. O _ _ _ _ _ !_ _! t I r-__ //////////////////// / // 33EA,33ED / // / // AV3, b / / '/! // ////////////////// CONNECTION TESTS - Fl(;UHF, 4-9

e bf AJfA k MMTsOA4GT6k a .aii,,,,,,,,,,. M r.-mm.. _ i.. e . ::::::: r i,.Ea r - =e. - --,.,n u. = n-2,- i.i. i6; h-g N wpouac cp., as w CON NECTION TEST SET-U P Figure 4-10 l l \\ 00

-? l 29 1, 1 . )4 4 1 2 25' LBS/FT OF CABLE LOAD N i 't D h to P l 8 l 5 12 a m) 8n g-- .&.INILOADED TRAY i l 8 e e s.1 e.2 e.3 e.4 e.5 c.s e.7 82 e.S 1A i

I Intef FleenSPECTRWIE2PAis)

Figure 4.11 - DAMPlNO AS A FUNCTION OF INPUT ZPA (Lower Bound) 9 e I 1

E m as ocH r .} / /- Y 4 w CLod- / EE ". ~E , WW l e a >o e k, me e cc n a L { wz38 M N w p-5 N wr mo o du 2 SE ^ ce ce 3 N o $w d 8 \\\\ a: W 1 x 8 u ~. N \\- u \\ T \\ \\ ( g*D ) NOI1YW31333Y

l '1 SR S R R T s es ~ P y<A f L j ^ f '0 /rY 3 -"6 G '1 s N2 a I w e PL e iE s a M. N D ) I G %D u / S 7L B ) P ] LE C OS SRS gt ( RT RNW )/ . CE O - L Y sv-C SS RR ( N TR i/ i. U 3 1 O e -,E 4 E / R R U / r G / I F w e e s e i e 70 " zO < - w J. w u* l;' i0 !

l: '

II

Ii' i

l! jIIlt llll1

e-e emm&+m e M N em e 4 4 8 g D C e. an 3ww E i C b U^ l LJ D zo e ZM e l c Cm Uw O M "O wJ 4a g nzwy 2 m o LsJ

m E W D

a3 g wcc r I g l 4 U2 C e D E. O l w N g R O .f, g m af M l I ( ot 3 3u - *O i=w l l i i ~

ANCC C3. ..... J,,,, _,, TES

fESG I

TESTOBECT Case 11 Strong Direction ACCE LOCAT.000 say#g D41g10/26/85 % X M wo 1"/mi t DATE 10/26/85 COeAE ifs A gy JCS gyg --- l jag 1615 gy JCS Y MTS Load Cell 20007/V<>1 t gyg_ --- g ___ 60.0 52.5 / \\ h 45.0 F \\ f I E I p O o 37.5 ~ x m + / 30.0 3 8 e S 8O 'I 22.5 O c; 15.0 y h m u g-4 p 7.5 1 0 .035 .070 .105 .140 .175 Rotation (rad)

I l ANC3 ENGINEET.S. lH3. 9887 JEFFET. SON CLVD, CULVE2 (,CA 300SS-hei g ygg TEST Alse TEST ORECT Case 11 Strong Direction ACCEL LOCAftDN ssW4 DAfg 10/26/85 m X M bt 1 % o14 DATE 10/26/85 N8 A gy_ JCS gyg 73g 1645 gy JCS Y MTS 1.oad Cell 2000#/Vo g --- g 60.0 W Y^ D 52.5 f a r 45.0 2 .c L E O A { 37.5 e T s 30.0 S E 8 E. B 22.5 .c> 0g 15.0 I T [ N '2 T 7.5 ? er O .035 .070 .105 .140 .175 o,... 4,,,, r,n

g g M M ANCC co. ..:.. =2, , am:.o .m. C=.m <.ca mv 7.4.1.1 3 Case 11 Strong Direction ACCS. LOCATIOtt mW/t 94Tg 10/28/85 peggsygag 10/28/85 gg genneggg a X celesco 1"/ volt b .ics 3,, l ,ygg1035 gy JCS g Y MTS Load Cell 20005/vol g g i 60.0 I i S2.5 u t.

1 T

l s i O e g S l a 37.5 x m N l 30.0 7 r 8 u S 8 E 8x 22.5 m>Om 7 15.0 c 2 e Q 7.5 a i n O 0)$ .070 .340 .375 Rotatinn (rad)

.1'-(l To k eduTAIUMEAIL_ 1 _ 4 D 3 % w m.u

:

M.,Qt iD i ..A \\Y i c... S-j g\\ e ~s \\ O \\ C ) V/

6. ' -

6W-17 Sw 18 4. o 9 0 b.,

g PMilAt ptAAl c/

.9 4,

S e,,

~

p GTAmmear BU4 tLL-)26c 0 (REF: DWq F-Slofo 7 REV 2o) 4, o .9 g a. f / FIGURE 4-15

~.P.=$

~ CF.ECKER IS CERTIFIED TO INSPECT Ohi' :. TRAY SUPPORTS CA3LE TRAY SUPPORTS PER ESG-l. - 2.5-CONSTRUCTED - O B1X,.bAlgm rat EL.d2C o s.' g,c c o., = =, CERTIFIED AS-CONSTRUCTED BY: ...g. GL 9 /baA/u o - ems 1 issue a ni u-8,* ' c,7 9763-L-370 /3 7 SN.2 DATE .c. o..: e g i:.,.,,3., e

t , fyjhy; - i.i O / EL. O'- 0" la'. a ' ;*,- 2'- 8" P' .$.h.* a.t-\\s'.o= ]. b$.4.' / pr ,3 1 A L___. q Jf I, b IZ rgAy I'. y [* _53 I r l i o i T i i @ _ _ _ _ - _5 r 7 n 33M I T1P l um I O l h I I I I

E _ _ _ _ _ _- l o i

F r i J. .' 4 0-l4 EMBEp 33AC 4 - e e, l . e.. s.. bE&lOM : be - 23 TA: (,309(o FIGURE 4-16 W.P.#7 CHECKER IS CERTIFIED TO INSPECT CABLE TRAY SUPPORTS O CAlLE TRAY $UPPORTS PER ESC-4 - AS-CONSTRUCTE BLDG (cuTAdvidI E1.{ t CERTITIED AS-CONSTRUCTED BY: ..........u,... NA ///3[ffo 0 is* FIRST ISSUE Jn)t )# U" '~' DAT' 9763-L-370 /3 7 50 32

+'4 Me @hwWh u+ eep et.=64 --'=b -6eMeq.

AM"wt

'%'tP'- 9 N O k e I i ? I /p,g y G. 4 60 n C,a 1 g.. i. l I l I I = Omas.*. 01 Cpes t (4) t=or-- iso o l - - 1 Il 1 I I i, t I i .0cq - kOf AT8 M h "d 01 j r Figu et 4 - \\~l - Tz e i c A t-Coeuccv w Bc a vi x 1

5.0 SUPPORT QUALIFICATION CRITERIA 5.1 Criteria Summary The purpose of this evaluation criteria is to provide the means by which to document the ability of cable tray supports to maintain their structural integrity during postulsted seismic loadings. Representative full-scale support testing (Reference 10) has demonstrated that adequate primary connection integrity ensures the cable's functionality and the support's ability to carry the cable load during and after a seismic event. Support behavior consistent with the full-scale tests is considered acceptable performance. The criteria is applied to supports and support features which are compatible with the full scale tests results. Specifically, the full-scale Seabrook tests (Reference'10) and the Bechtel Generic Raceway tests (Reference 6) have demonstrated that typical cable tray support systems behavior is essentially controlled by the behavior of the primary overhead (vertical load) carrying connections and the brace connections. Supports of compatible geometry and hardware with the full scale tests, then, can be qualified utilizing this criteria. Implementation of this criteria, consistent with the test results, will ensure acceptable performance of the cable tray supports consistent with the full scale test results (Reference 10). primary load path stresses will be limited to levels identified in Section 5.4. Figure 2(b) serves to define support primary stresses and test compatibility. Case I is completely test compatible, based on its similarity to the Case A test presented in Reference 1. The vertical strut and primary connection create the primary load path, and require evaluation. The brace and its connections are not primary, and are more fully discussed in Section 5.4. Full scale testing has demonstrated the brace is secondary in terms of support integrity and functionality. Figure 2(b), Case II, is compatible with the test program, however, its geometry creates additional primary load paths (and stresses). Cable load is carried both horizontally.

. ~...-...... -.., and vertically. Figure 2(b), Case III, is also compatible with the test program. Supports and/or support features not compatible with the test (i.e., welded steel frame support) will require a unique support evaluation. These supports will be qualified in accordance with Reference 11. pertinent test data (e.g., connection test results) can be used to enhance the structural modeling and evaluation of cable tray supports, as applicable. A flow chart depicting the various evaluation alternatives is presented in Figure 1. 5.2 Scope The scope of this evaluation criteria can be applied to the cable tray support system as follows. The cable tray evaluation is limited to members, connections, etc., whose sole purpose is to support the cable tray. Figure 2(a) presents a typical cable tray support which serves to clarify the scope of evaluations performed per this criteria. Specifically, the limits of this criteria are as follows: Supplemental Steel - Reference 11 provides the technical guidelines for this evaluation. Support Anchorages - System anchorages (embedded plates, inserts (including braces) and chennels; surface plates and channels, welded channels) are within the scope of the evaluations performed per this criteria. Specific anchorage evaluations should be performed in accordance with Reference 11. primary Connection, etc. - The support proper (i.e., primary connections, brace connections, etc.) are to be evaluated utilizing this criteria, as outlin'd in Section 5.1. e _

j i 5.3 Evaluation by Test Support evaluation by test is limited to configurations where critical elements of the support under consideration are enveloped by the tested conditions. The four major test configurations are depicted in Figures 3 through 6; however, test details are provided in Reference 10. The cable tray suqport can be qualified directly utilizing the system test data with proper consideration of the following parameters: o Support Width o Support Height o Support Boundary Conditions o Cable Mass o Hardware Compatibility o Bracing o Span o Response Spectra planar (i.e., transverse and vertical) enveloping is assured if, individually, all the above parameters are enveloped by a tested support. Further, planar enveloping can be demonstrated for any configuration where the critical primary elements (e.g., the primary load paths) of the support under consideration are enveloped by the tested conditions. l l In addition to enveloping the "in-plane" geometry, consideration must l be given to the "out of plane" (longitudinal) dimension effects, that is, the routing of the overall cable tray system. Case A and B, as shown in Figures 3 and 4 have a general shape (in plan) of the letter "J". This "J" - -.

configuration was chosen to investigate the load distribution of the global "X" direction (longitudinal). The one support "around the corner" (i.e., the support perpendicular to the other four support:) served to stiffen the longitudinal system direction, and typically must be considered in any "out of plane" evaluation. For qualification purpose, then, the system restraint in the longitudinal direction requires evaluation. Longitudinal restraint can be provided by a combination of support framing, system restraint, and/or brace resistance. To carry the longitudinal load, direct longitudinal support of the system can be provided, supports perpendicular to the longitudinal direction may be used, or adjacent supports can be braced together as a means of accepting the lateral load. The "J" configuration used in the test survived both braced and unbraced conditions; therefore, various in situ conditions (and brace patterns) can be qualified by direct comparison to the test program. Routing configurations which provide additional support to the tray system can be accepted by test. The overall eccentricity of the system should be considered. The tested "J" sample was subjected to overall eccentricity (Figures 3 and 4). Geometries with greater eccentricity (due primarily to larger unbraced lengths) must be considered. 5.4 Evaluation by Combination Test and Analysis It has been demonstrated by the dynamic r, stem tests (Reference 10) and the capacity and fatigue connection tests (References 10 and 12) that the critical elements for support and system integrity are the primary overhead connections. Correspondingly, the analyses of test compatible supports will be directed towards the investigation of the above key parameters. Supports qualified by test / analysis will require the primary load paths to be verified for integrity. This will be demonstrated by the following:. - -

Primary Connections - These connections must be demonstrated to adequately carry the vertical load during and after the seismic event. Primary connections subject to joint rotations due to seismic loads must be evaluated to verify the connection's ability to carry the primary loads per Section 5.4.1. Brace Connections - These connections must be demonstrated to maintain integrity during and after the seismic event. These connections must be evaluated per Section 5.4.2. Displacements - Support displacements may require limitation due to their immediate proximity to other plant hardware. The local proximity must be evaluated for available clearances. In these situations the support displacement must be limited by existing hardware, or the support will be modified to strengthen its lateral support and/or longitudinal restraint. 5.4.1 Primary Connections A sample trapeze type support is pictured in Figure 7. Support flexibility is controlled by the rotational stiffness of the primary connections and the translational stiffness of the brace hardware. These key features are then modeled in any support mathematical evaluation. A typical test moment-rotation diagram for a primary connection is provided in i Figure 8. Mathematically, this behavior can best be represented by bilinear elastic-plastic behavior. The connection, although possessing some initial l stiffness, will essentially rotate at a constant moment at higher loading conditions. To ensure structural integrity, primary connection rotation will be limited to its ultimate value derived by connection testing and divided by a factor of (to be established after completion of testing). Conversely. l L

the connection may be conservatively modeled as a simple pin and the joint j rotation must meet the acceptance criteria. ) Primary connections must be evaluated against the allowance derived per Reference 11 in the load carrying direction. Connecti'on tests (still in progress) will be utilized to develop these interaction curves. e 5.4.2 Brace connections Brace connections must also maintain their integrity. They differ from the overhead primary connections, however, as they may or may not function as primary connections. Their role as a primary connection is dependent upon their application. Brace connections will be considered a primary connection if they are utilized in the following applications: (1) Serves as a wall support - the connections must carry vertical and horizontal loads (refer to Figure 2(b)); (2). serves to limit the rotation of primary overhead connections to within acceptable limits; and (3) serves to limit support displacements to prevent impact on other plant hardware. For these applications, the brace connection will be limited to its allowable load. Brace connections which are not considered primary connections do not need to offer resistance. Integrity of the brace will be maintained if the brace is limited to 70 percent of its ultimate displacement derived by test. A typical load deflection curve for brace connection is presented in Figure 9. I 5.4.3 Modeling Methodolomy/ Evaluation Analytical modeling will utilize classical structural modeling i techniques, similar to those in Reference 11, with a few notable exceptions. Connection stiffness, based on test program data, can be inserted into l l the mathematical models in the form of translational and rotational springs. l The dynamic system tests de.nonstrated the importance of the connections in l t 1 l

determining system dynamic properties (frequencies, mode shapes, etc.) and system fragilities. The actual spring rates to be modeled will be based on the connection test data performed at ANCO (Reference 10) and the data sununarized in Reference 12. The evaluation of these key connections is summarized in Sections 5 4.1 and 5.4.2. Modeling/ analysis will essentially fall into one of two categories - system analysis and supplemental analysis. A system analysis consists of the creation of a complete analytical model for a single cable tray support or a group of supports and cable tray. The model sust be large enough to reflect the load transfer of the support system. The response levels extracted from the system analysis are then used to evaluate the primary stresses required by this criteria. Supplemental analysis can be performed without a system analysis or finite element model development. Supplemental analysis is used to evaluate minor deviation between in situ and tested support systems. Typically, this would encompass evaluation of a unique connection, etc. System analysis, in general, requires the use of either static or dynamic analysis techniques. These procedures are described in Sections 5.4.3.1 and 5.4.3.2. 5.4.3.1 Static System Analysis A static system analysis can be performed to qualify a cable tray support system meeting the general requirements of Section 5.1. If the static approach is used, the analysis must conform with the requirements of Reference 11 with the following exceptions: a. Modeling (and analysis) is not limited to the use of the STARDYNE computer code. Other verified finite element computer codes may be utilized. l __

b. Damping values utilized must be consistent with Section 5.5 of this criteria. c. Consistent with Section 5.5 of this criteria, all analysis will be performed using the SSE condition seismic loads. d. The evaluation requires an assessment of the primary connections as e established in Section 5.1. 5.4.3.2 Dynamic Analysis Method The dynamic analysis can be performed using a modified approach to the step by step procedure outlined in Section 6.2.2 of Reference 11. The initial six steps in Section 6.2.2 relate to specifics of the STARDYNE approach and, therefore, would be modified to utilize any alternate computer program. Step 7 would utilize Steps 11 through 19 of the static load approach. Rather than tie the dynamic analysis approach specifically to a computer code, the following guidelines are provided to develop the analytical model. Acceptable alternates may be utilized provided that adequate justification is presented. a. Model size - The model must be sufficient in size to accurately account for system load transfer. b. Cable tray ends (at the midspan between supports) are assumed to be free in translation and fixed in rotation to simulate system continuity, when applicable. c. Significant mass which has been distributed by the computer code to nodes which have not been assigned dynamic degrees of freedom is lumped from such nodes to the nearest node which has been assigned a dynamic degree of freedom, d. The support system must be evaluated to support the mass of the cable, cable tray, support members, miscellaneous hardware, and attachments (wireway, etc.). A contingency for future cable additions is normally included.

Modal combinations (using the response spectra techniques) should e. be by the SRSS technique; however, all modes within 10 percent of each other will be combined by absolute sum, in accordance with Regulatory Guide 1.92. Unless justified, a minimum of 85 percent of the model mass must be retrieved in all three directions at frequencies up to 33 Hz. If this mass retrieved in any direction is less than 85 percent, then the effect of the balance of the mass (i.e., the " rigid" mass) can be added as a static inertial load by using the floor ZPA. Cutoff frequencies less than 33 Hz may be utilized, when appropriate. f. Directional combinations (of the two horizontal and one vertical direction response) should utilize the SRSS technique for response spectra solutions, s. Time history analysis should be in accordance with the guidelines of Regulatory Guide 1.92. h. Items 4.3.1(i) through 4.3.1(iv) apply to all dynamic analyses as well as the static analyses. 5.5 Required Review Data 5.5.1 Loads Loads should include weights of the following, whichever is applicable. a. Cable l b. Cable Tray and Tray Covers c. Support Members The load for any support should be derived for all tray levels in a consistent fashion. That is, the maximum loads in Reference 11 can be used l unless the loads are selected to more closely bound the in situ conditions. i 1 __

However, the system under evaluation must utilize either entirely full design loads or in situ loads, exclusively. 5.5.2 Damping Damping valves will be in compliance with Figure 10. D As shown in Figure 10, the design damping for Seabrook cable tray supports is dependent upon: a. Cable Load b. Input Floor Response Spectrum ZpA For trays loaded with between 25 lb/f t and 35 lb/f t of cable load (i.e., total cable plus t ray load of between 30 lb/f t and 40 lb/f t), the appropriate damping values range from 7 percent and 20 percent, depending upon the input floor response spectrum 2pA. For trays without cables, damping values consistent with USNRC Regulatory Guide 1.61 are applicable. For cable loa. ling less than 25 lb/f t, linear interpolation is used to determine the applicable damping value. 4 - - - - -

i i SUPPORT AS-BUILT DATA TEST DATA J S THE SYSTE I ENVELOPED BY AN YES - EXISTING QUAL'N' NO i l IS THE SUPPOR . _ NO SYSTEM UNDER REVIEW ' TEST COMPATIBLE'? YES l CAN THE SYSTE. '- NO BE QUAL'D DIRECTLY YES BY TEST 1 QUALIFY SUPPORT PER REF. 11 I PERFORM PERFORM ANALYSIS PER TEST COMPARISON SECTION 5.4 PER SECTION 5.3 f DOCUMENT QUALIFICATION ~ ~~ ~- - -- -~ l l l SYSTEM QUALIFICATION FLOWCHART l FIGURE 1 [

f SUPPLEMENTAL STEEL e--MAIN STRUCTURAL STEEL b,, L Qp V I i f I / nfk) (2 BRACE ANCHORAGE INTERNAL __ g \\ CONNECTION t 9 (TYP) BRACE CONNECTIONS. TYP + TYPICAL CABLE TRAY SUPPORT FIGURE 2 (a) 1 I e I

PRIMARY CONNECTION (TYP) 4L L VERTICAL v T STRUT. TYP y y I I ' 4-BRACE CONNECTION 77p ~~ ( HORIZ. STRUT, TYP _\\ STRUT CD:;WECTION PRIMARY CONNECTIONS 40 4L 4'.k' c u s 1 1 1 I BRACE CONNECTION - TYPICAL CABLE TRAY SUPPORT CONFIGURATIONS FIGURE 2 (b) l n- . - +

l 3 l j l 4 i j TRAY t I 4 \\ L MV I I AY Eiur LA 1 RA A ew i -k [M M RB 77 3 VA I vs i "fk-1 -t, \\ t i q' = \\ i-1, CASE A i I l l' FIGURE 3 l l l t i~ -.

I t ? m.L13' l l f l N g 4 ,.1:: m W h W 2 l A 1 l n 1 1 s '[ Ik AWf ' g C L 1 W l e,------.------r-,,,,,-,,,,_,w,-,,,-nw ,-w,_,,.,-n,,

--m- " " ^ - " " - - - - ' D 9 O gs u i W I .m (v i 11 o 4 w i L1J E d i V) 2 I 6 15 -3 Q .'s %T@L s isii l t -..----w .n -- n,-_,_,

e g ~ ~ I I-g g3, o g / / // / m w e.. E' E Q ( x Y I

,------,e-,,

a-

!-PRIMARY CONNECTIONS sti t ve 99 gry BRACE CONNECTIONS I l, ri ga. Ik 1 ,i r TYPICAL CABLE TRAY SUPPORT l FIGURE 7 f l l l l

i;, I s D A \\ n (R J

i. N

'T O R A I 0 s . m_ v M U 8 LD E R A 3 UG C3 I F I P / Y T / / / o. s. 7 n i mE,s nzo2o= t l ii

.. ~... _ e AXIAL LOAD (P) m r DISPLACEMENT (D) TYPICAL LOAD DEFLECTION CURVE FOR 1 BRACE CONNECTION FIGURE 9 1 \\

1 N i 2: 25' LBS/FT OF CABLE IAAD = N l e A ) N i n 9 I e g 2 i e 09 + 8 pg f- .WWLSAste.T.R.AF e l ? 4 9 O S.1 82 R2 9.4 9.5 0.0 EF GA GA 13 INPST PLSSRWetTWWWIFR W Lower Bound Desping as a Function of input ZPA Note: For tray loaded less than 25f/ft (cable), linear Interpolation is used to determine the associated design damping value. ~ FIGURr 10 i l l

6.0 SEABROOK CABLE TRM StFPORT EVALUATION PROCEDURE This procedure sismarizes the acceptance criteria for cable tray supports at Seabrook. A ntaber of full scale tests of cable trays and supports were conducted. These tests demonstrated the outstanding

integrity of the existing cable tray and support configuration and provide the basis for the criteria.

The detailed test procedure and results are provided in Reference 10. The following criteria is to be utilized in the categorization, evaluation and acceptance of cable tray supports and is based on the Evaluation Criteria defined in Section 5.0 of this report. 6.1 APPLICATION In the process of categorizing, evaluating and accepting cable tray supports the use of check lists is required. Rather than allow the user to interpret the Evaluation Criteria on an item by item basis, this document provides the detailed acceptance criteria and presents the background and technical justification for each step in the process. For each support the check list will require reference to the appropriate sections of this report which are used in the categorization, evaluation and acceptance process. See Figure 6.5 for a sample check list. 6.2 Categorization In order to determine the type of evaluation to be performed on a specific support the check list user must first determine the categorization status of that support. Two categories are provided as follows: l 34

6.2.1 Test Configuration Figure 6.1 provides support geometry details that are qualified by the dynamic tests. In addition to assuring that the support under consideration complies with one of the details in Figure 6.1, the evaluator must assure that the cable tray section containing the support satisfies the displacement criteria of Section 6.4. The support under consideration is acceptable if the dimensions and loads are equal to or less than those provided by Figure 6.1 and the cable tray geometry complies with Figures 6.2(a), 6.2(b), 6.2(c) and 6.2(f). 6.2.2 Analyzed Configuration It is anticipated that individual supports may exist which are not covered by the Test Configuration (6.2.1). For such cases an evaluation in accordance with Section 5.4 is required. In order to reduce the number of individual support calculations required a ntaber of generic analyses have been performed. These analyses include consideration of the cable tray system as well as an evaluation of individual supports. j The support under consideration is acceptable if the dimensions and loads are equal to or less than those provided by Figure 6.3, if the support exists in a cable tray geometry in compliance with Figure 6.2 and l l the support is typical of an analyzed model, e.g. Figure 6-2(d) or Figure 6.2(e). 35 l l

6.3 ACCEPTAf1CE CRITERIA In the process of categorizing supports, the evaluator automatically adopts an acceptance criteria when the specific support f alls into th'e Test Configuration (6.2.1) or the Analytical Configuration (6.2.2). In order

to provide technical background and justification for that acceptance the following discussion of each category is provided.

6.3.1 Test Configuration Supports which f all into the Test Configuration category are acceptable because they are represented by the supports subjected to dynamic testing which satisfied the following limits (Reference 10). 6.3.1.1 Loading The section of 24" wide cable tray, including support, which was tested had the following loading: Tray Fill 40% (25-50 lbs/ft) No. of Trays 6 Max. Support Spacing 10 feet TRS Peak (SSE) max 7.8G 0 10Hz 5.5G Fatigue 5 OBE + 1 SSE Max TRS Peak 9.3G to 11.7G 36 j

1 6.3.1.2 Resultant Stresses The maximum peak strains measured in the support i members results in the following stresses for the case with braces present. TRS Peak (SSE) 2.7 KSI Max TRS Peak 3.9 KSI The tests were rerun with all bracing removed and the TRS (SSE) was applied. The maxim:sn member stress was 3.6 KSI. 6.3.1.3 Cable Tray Displacements The displacements of the cable trays during testing is as follows: Case A (SSE) Braced Longitudinal 4.29 Inches Lateral 3.81 Inches Unbraced Longitudinal 5.12 Inches 1.2 SSE Lateral 5.12 Inches 1.2 SSE Case B (SSE) Braced l Longitudinal 1.19 Inches Lateral 3.85 Inches Unbraced Longitudinal 1.96 Inches Lateral 4.63 Inches, i 37 l l

Unbraced Trapeze (Floor Connections Removed) longitudinal 3.35 Inches Lateral 5.62 Inches Case C (SSE) Braced Longitudinal 0.79 Inches Lateral 1.30 Inches Unbraced Longitudinal 1.46 Inches Lateral 2.82 Inches 6.3.1.4 Sunnary The loadings applied to actual cable tray supports 4 which are represented by the Test Configuration are less than those applied in the test since the actual fill is equal to or less than 40% for all trays, the support spacing is normally less than 10 feet and the maximum peak acceleration from any amplified floor response spectrum that is applicable to cable trays is 6. Using a static equivalence approcch the I load per support can be determined as follows: Test I F = 10 ft. (40 lbs/ft) 7.8G = 3120 lbs. As-built (typical) F = 8 ft. (20 lbs/ft) 6.0G = 960 lbs. Generally, then, the as-built support is approximately 36% that of the test support load. 38 l ~

Based on the above, any support satisfying the requirements of Section 6.2.1 will have lower loads and stresses than the tested j support and is acceptable. 6.3.2 Analysis Confis. ; tion Supports which fall into the Analysis Configuration category are acceptable because they are represented by an analysis of a cable tray and support geometry that is similar in load and acceptance criteria. A set of generic analyses have been performed. These analyses represent cable tray routing and support geometries exemplified in Figures 6.2(d), 6.2(e), 6.2(g), 6.2(h) and 6.2(i). The analyses considered actual cable tray loading, envelope amplified floor response spectra, appropriate damping and connection stiffness. The resulting displacements (including connection rotations) and primary member stresses satisfied the requirements of Section 5.0. Displacements were evaluated with respect to cable terminations and adjacent components or structures. It is anticipated that a significant ntsnber of supports at Seabrook will fall into this category since general support geometries are [ quite similar. Major differences are related to cable tray routing, number of trays, cable fill per tray and allowable displacement. i l 39 1 l l

6.3.3 C00stECTION DETAILS An integral part of the acceptance of a support is the connection details. Primary connections, which are those connecting the support to building structure, and Secondary connections, which are those 6 connecting members of the support together, have been qualified by testing. However, the evaluator of the support must determine that the connections are acceptable. This is accomplished by assuring that all connections are in compliance with Figure 6.4 which indicates acceptable details along with allowable loads. The allowable loads on the connections have been determined from testing and include load cycling effects. It is important to recognize that cycling effects are not a requirement for design of cable tray supports, however, the allowable loads of Figure 6.4 include this effect. The cyclic effect is included since Seabrook test indicated degradation of 33EY, EC, EA and ED brace connections which results in higher loads on primary members and tray displacements larger than those which would be determined assuming full capacity of those connections. 6.4 DISPLACEENT CRITERIA 6.4.1 Analysis Configuration Supports which f all into the Analysis Configuration category are acceptable because they are represented by an analysis of a cable tray' and support georoetry that is similar in load and acceptance criteria. A set of generic analyses have been perfonned. These analyses represent cable tray routing and support geometries shown in Figure 6.2(d), 40

6.2(e), 6.2(g), 6.2(h) and 6.2(1). The analyses considers actual cable tray loading, envelope amplified floor response spectra, appropriate damping and connection stiffness. The resulting displacements (including connection rotations) and primary member stresses satisfied the e requirements of Section 5.0. Displacements were evaluated with respect to cable terminations and adjacent components or structures. + t i i l i 41

TT ll a ga i i __ N I i A N-k 1 Ni i i N -) - p LB-f a 3 r -S-DMD 2 3 1 Figure A B N Q LOAD la Ib NOTES:

1.. A trace is not required for acceptance.

2. Di;Nacement criteria of section must be satisfied. 3.

.- r, umber of trays.

I e FIGURE 6.1 42 m. m

__g _ S. 7 i I l i l t _J _j k i E t E (a) (b) (c) =-4=. . T t, I l i i i I I l .p. ..1-s (d) (e) (f) ~4r -sp. --1 7-N T~~ l i i I i l I I i ~T 4-L4r- - a -. .-N-i (g) (h) (i) l FIGURE 6.2 43

-m

---a*E+W--h w se h-e4+

m ep A '%e V [ I g I i I g i t a i_. g A ~ i 1 k i i D L_ _ _ _ J I i I y %S% B-I c = (a) (b) t l l i, A L_._- i I i i I L I I y B -. B E D =

== (C) FIGUP.E 6.3 44

FIGURE 6.3 (continued) DEAD 3 1 Figure A B C D E N e LOAD 2 3a NA ,g gg 3b NA NA tm 3c NA la FM NA 3d NA 3e NA NOTES: 1. A brace is not required for acceptance of Fiqures 6.3(a) and 6.3(b). 2. ta = Not applicable. 3. N = Number of trays. 9 f 4 0 e i 45

e . /////////// l \\ 1 l l s. i i l ( ) ( ) ( ) ( ) 3300 33DC 33FN 33FD PRIPMRY CONNECTIONS l / # ~ 30* j i i i i ( ) ( ) ( ) ( ) 33EC/33EY 33EA/33ED 33DC 33DF BRACE P1E!!BER CONNECTIOMS l l I 1 i 1 l i e i ( ) ( ) ( ) 33DA 33DQ 7-CLIP TRAY SUPPORT CONNECTIO" NOTE: 1. ( ) Allowable connection load, t FIGURE 6.4 46

SEABROOK CABLE TRAY SUPPORT QUALIFICATION CHECK SHEET Support I. D. Number: SUPPORT CONFIGURATION PARAMETERS (fill-in) a. Support height, A = b. Support width, B= c. Support brace angle. 0 = d. Number of trays, N = y(,,x), Qualified By e. Section I. SUPPORT CATEGORIZATION (check one) 1. Test Configuration 2. Analyzed Configuration 3. Analysis Required Configuration II. SUPPORT LOADING PARAMETERS (fill-in) a. Number of trays: b. Max. width f trays: in, c. Max. % tray fill volume: d. Ave. support spacing to adjacent supports: ft. e. Peak floor response spectra acceleration: Horiz. G's Vert. G's III. SUPPORT BRACING PARAMETERS (check one) 'a. Lateral bracing None: Horiz: Diagonal: b. Longitudinal bracing None: Horiz: Diagonal: l IV. SUPPORT CONNECTION PARAMETERS (identify) 'a. Primary connectior s b. Brace member connections c. Tray support connections V. SUPPORTDISPLACEMENTPARAMETERS(identify) a. Distance to cable termination point: in. I b. Distance to adjacent components: in. BY: DATE: i CHK'D BY: DATE: l l FIGURE 6.5 L os

i

7.0 REFERENCES

l 1. Letter from J. DeVincentis (PSNH) to C. W. Knighton (NRC), " Cable Raceway System Damping," dated June 3, 1985. 2. " Test Plan - Performance Testing of a Typical Cable Tray Configuration: Seabrook Station - Test Cases A & B " ANCO Engineers, Inc., Document A-000146, Prepared for Bechtel Power Corp., August 1985 Revision 1. 3. " Test Plan - Performance Testing of a Typical Cable Tray Configuration. Seabrook Station - Test Case C," ANCO Engineers Inc., Document A-000151, Prepared for Bechtel Power Corp., September 1985, Revision O. 4. " Test Plan - Performance Testing of Typical Cable Tray Support Connections - Seabrook Station," ANCO Engineers Inc., Document A-000147, Prepared for Bechtel Power Corp., September 1985 Revision O. 5. "Sunmary Report - Cable Tray Support Damping at Seabrook Station," Bechtel Power Corp., June 1985. 6. " Cable Tray and Conduit Raceway Seismic Test Program - Release 4 (Final)," Test Report No. 1053-21.1-4, Volumes 1 and 2, December 15, 1978, Volume 3. May 1980; Volume 4. March 1981, ANCO Engineers Inc. 7. Meeting Summary, Prepared by V. Nerses, USNRC, Applicant - PSNH, Facility - Seabrook Station, Units 1 and 2, dated June 25, 1985. 8. Safety Evaluation Report, NUREG-0881 Docket STN 50-482, USNRC, PP. 3-11, dated December 1981. 9. Safety Evaluation Report, NUREG-0831, Supplement No. 1. Docket Nos. 50-416 and 50.417. USNRC, Section 3.7.3, December 1981, 10. " Performance Testing of Typical Cable Tray Configurations and Strength Testing of Selected Connection Details, Seabrook Station Test Cases A, B, C." Test Report No. 1053.43B, Document No. A-000161, ANCO Engineers, i Inc. 11. " Technical Cuide for the Design and Analysis of Seismic Category I Cable Tray Support Systems," PSNH, Seabrook Station, Revision 2. March 6, 1984, PIN No. 9763-SQ-00121, 82025. l 12. " Strut Nut Connection Test Report - Phase II, PSNH Seabrook Station, Units 1 and 2," Report No. 9763-SQ-00121-32014, June 9, 1983. i l i' i 48 i

APPENDIX A 4.0 A COMPARISON OF CABLE TRAYS AT THE SEABROOK NUCLEAR STATION W THE SEISMIC EXPERIENCE DATA BASE The seismic capacity of cable tray systems at Seabrook Station is assessed through a study of comparable cable tray systems that have experienced strong motion earthquakes. The excellent performance of cable tray systems that have experienced seismic motion in excess of the Seabrook design-basis safe shutdown earthquake (SSE) provided the basis for this study. The experience of cable tray systems in past earthquakes is extracted from the data base of earthquake effects to power and industrial facilities, compiled by EQE Inc., under the sponsorship of the Seismic Qualification Utilities Group (SQUG). This experience data base, encompassing approximately 20 facilities and a total of eight major earthquakes, includes a diversity of cable tray systems. The total data base inventory includes on the order of 10 miles of cable trays which experienced strong motion earthquakes. This inventory of data base cable trays covers a wide range of structural types, support configurations, cable tray system layout, locations within buildings, and seismic input loads. Most of the sites surveyed in compiling the data base experienced ground motion comparable to or in excess of the Seabrook Station safe shutaown earthquake. Within the experience data base there is only one instance of seismic damage to a cable tray structure. This single instance of seismic damage was to an exceptionally weak, atypical cable tray support structure subjected to high seismic load. Normal industry practice in the construction of cable tray systems does s not include a specific design for seismic loads. Cable tray supports structures (outside of nuclear plants) are normally designed to accomnodate gravity load only, in spite of this fact, the performance record of cable trays in past earthquakes is excellent. Cable tray systems display a large margin for the absorption of seismic loads without damage. This inherent seismic margin in cable trays does not appear to be sensitive to variations in cable tray construction, layout, or seismic input.

e' J The purpose of this comparison of cable tray systems in Seabrook Station with cable trays in the seismic experience data base is to illustrate the following points: The seismic experience data base includes all parameters a associated with the ability of cable trays to resist seismic loads, The parameters associated with the seismic capacity of data a base cable trays envelop the parameters of Seabrook Station cable trays, i.e., data base cable trays range from similar to weaker when compared to Seabrook cable trays, In most cases data base cable trays have experienced e seismic loads comparable to or in excess of the design basis seismic loads for Seabrook cable trays. Cable tray systems constructed according to normal industry m practice, even without specific provisions for seismic loads, are sufficient to withstand earthquakes in excess of the seismic design basis for Seabrook Station. The diversity of critical parameters encompassed by the e cable tray data base demonstrates that the capacity to survive earthquakes is not sensitive to minor variations in the details of cable tray construction or layout. For this reason cable trays are generically adequate to sustain moderate seismic loads (such as the design basis for Seabrook Station), as long as their construction conforms to or exceeds normal industry practice. 4.0-1 The Seismic Exoerience Data Base Details of the performance of power and industrial facilities in past earthquakes have been compiled into a seismic experience data base by EQE. The primary sponsor for the compilation of this data base has been the Seismic Qualification Utilities Group (SQUG). The SQUG was organized in 1981 by a group of electric power utilities with operating

nuclear plants. The primary purpose for the organization of the SQUG was to develop a program to address Unresolved Safety Issue A-46, the potential seismic hazard to critical equipment in operating nuclear plants. The basis fo-the SQUG program was to determine the realistic seismic hazard to the equipment installations of nuclear power plants, based on the experience of facilities with comparable installations in past earthquakes. Very few components of critical nuclear plant systems are specific to nuclear facilities. Critical nuclear plant systems include electrical switchgear, control panels, motor-operated valves, pumps, { piping, ducts, conduit, cable trays, etc., all common components of conventional power plants and large industrial facilities. Strong motion earthquakes frequently occur in California and Latin American Countries, where power plants or industrial facilities are included in the affected areas. By studying the performance of these earthquake-affected (or data base) facilities, a large inventory of various types of equipment can be compiled that have experienced substantial seismic.iotion. The ground acceleration experienced at most data base sites, measured by nearby ground motion records, is comparable to or in excess of the seismic design basis for most eastern United States nuclear plant sites. The primary purposes of the seismic experience data base are summarized as follows: To determine the most common sources of seismic damage, or a adverse effects, on facilities that contain installations representative of critical nuclear plant systems. To determine the thresholds of seismic motion corresponding to various types of seismic damage, To determine the types of installations that are normally e undamaged by earthquakes, regardless of the levels of seismic motion. 1

To determine minimum standards in equipment installations, e based on past experience, to assure the ability to withstand anticipated seismic loads. To summarize, the primary assumption is that the actual seismic hazard to nuclear power plant installations is best demonstrated by the performance of similar installations in past earthquakes. 4.0-2 Facilities Surveyed in Compilina the Data BAlt The seismic experience data base is founded on studies of over 60 facilities located in the strong motion areas of 10 earthquakes that have occurred in California and Latin American countries since 1971. The data base was compiled through surveys of the following types of facilities: Fossil-fueled power plants e Hydroelectric power plants e Electrical distribution substations e Oil processing and refining facilities e Water treatment and pumping stations a Natural gas processing and pumping stations a Manufacturing facilities e Large commercial facilities (focusing on their HVAC e plants). In general, data collection efforts focused on facilities located in the areas of strongest ground motion for each earthquake investigated. Facilities were sought that contained substantial inventories of mechanical, or electrical equipment, or control and instrumentation systems. Because of the number of earthquake affected areas and types of facilities investigated, there is a wide diversity in the types of installations included in the data base. For the equipment 4 -____f. ..m__-

Installations of focus in the investigations, this means a wide diversity in age, size, configuration, application, operating conditions, manufacturer, type of building, location within building, local soll conditions, quality of maintanence, and quality of construction. The data base includes a total of ten earthquakes, with several different sites investigated in each earthquake-affected area. The earthquakes investigated range in Richter magnitude from 5.7 to 8.1, Measured or estimated ground accelerattons for data base sites range from 0.15g to 0.70. 9 The duration of strong motion (on the order of 0.10g or greater) ranges from 5 seconds to over 40 seconds. Local soil conditions range from deep alluvium to rock. The buildings housing the equipment installations of interest have a wide range in size, and type of construction. As a result, the data base covers a wide diversity of seismic input to equipment installations, in terms of seismic motion amplitude, duration, and frequency content. 4.D-3 Tvoe of Data Collected 't Information on each data base facility, its performance during the earthquake, and any damage or adverse effects caused by the earthquake were collected through the following sources: Interviews with the facility management and operating e personnel usually provide the most reliable and detailed information on the effects of the earthquake on each facility. At most facilities several individuals were consulted to confirm or enhance details. In most cases interviews are recorded on audio tape. The facility operating logs provide a written record of the e conditions of the operating systems, before and after the earthquake. Operating logs list problems in system operation associated with the earthquake, and usually tabulate earthquake damage to the facility. Operating logs are useful in determining the amount of time the facility E li'

may have been out of operation followir.g the earthquake, and any problems encountered in restarting the facility. The facility management often produces a report summarizing e the effects of the earthquake follcwing detailed inspections. These reports normally desribe causes of any system malfunctions or damage, and typically include any incipient or long term effects of the earthquake. If the facility can be surveyed immediately following the m earthquake, as has been the case in four of the ten earthquakes included in the data base, earthquake damage can often be inspected prior to repairs. Standard procedures used in surveying data base facilities focus on collecting all information on damage or adverse effects of any kind caused by the earthquake. For a large majority of the facilities surveyed in the data base, this is not a lengthy task. Except for sites that experienced very high seismic motion (in excess of 0.50 g peak ground acceleration), seismic damage to well-engineered facilities is normally limited to only a few items. 4.0-4 The Data Base for Cable Trays Within the experience data base, approximately 20 facilities, encompassing eight earthquakes, include good examples of cable tray systems. Table 4.D.1 lists these facilities with a brief description of the type of cable trays found at each site. In general the data base offers a wide diversity of cable tray designs, configurations, locations within building structures and conditions of seismic loading. Figures 4.0-1 through 4.0 3 include illustrations of the primary parameters that affect the seismic loads on cable trays for several data base sites. These figures include a plot of the response spectrum representing ground acceleration at the data base site. This response spectrum is based on the nearest or most applicable ground motion records for the site. The response spectrum shown is the average of the response spectra for the two horizontal components of motion measured at i

r. ,. _., ~ - - the nearest record. This spectrum then does not represent the highest horizontal motion at the site, but rather the average horizontal. motion. For comparison, each plot also includes the horizontal ground motion spectrum from USNRC Regulatory Guide 1.60, normalized to a peak ground acceleration of 0.25 g, the design basis for the Seabrook Station. All response spectra correspond to 5% damping. Each figure includes a schematic elevation view of the data base building structure, illustrating the height and construction of the building, and showing the primary locations of the cable trays with respect to grade elevation. The figures also include sketches of the typical cable tray construction at the site, including the typical number of tiers, support configuration, and attachment to walls, floor, or ceiling. 4.0-5 Ca_ble Tray P_arameters at Seabrook and in the Ernerience Data Base In order to verify the seismic adequacy of the Seabrook cable tray systems using experience from past earthquakes, relevant parameters must be chosen which make comparisons between cable trays in the data base and at Seabrook meaningful. Cable tray parameters were defined based on their effect on system mass, stiffness, strength, and response to seismic loading. The following critical parameters are the basis of the comparison of cable trays at Seabrook Station with those in the seismic experience data base: Cable tray dimension (width, depth) a Cable tray loading e Number of tiers a Cable tray type (ladder, trough, solid bottom) a Support construction (trapeze, cantilever bracket, rod, e Unistrut) Cable tray span (length betw. ten supports) = t I

Connection details (e.g., tray-to-support connection) = Additional cable tray support loading (conduit, piping) Cable tray interfaces (with electrical cabinets, with e conduit) Cable tray layout e Location of trays e Type of building a Seismic ground motion a Each of the above cable tray parameters at Seabrook is compared to the seismic experience data base. Data for the parameter comparison were taken from the following sites, which provided the most cable tray details: Sylmar Converter Station (PGA - 0.50g) e Valley Steam Plant (PGA - 0.30) e Humboldt Bay Power Plant (PGA - e 0.25, 0.30g) 9 El Centro Steam Plant (PGA - 0.42g) = Drop IV Hydroelectric Plant (PGA - 0.40g). e las Ventanas Power Plant (PGA 0.30g) e la Villita Hydroelectric Plant (PGA 0.15 ) e 9 El infiernillo Hydroelectric Plant (PGA 0.15 ) e 9 Each of the critical cable tray parameters listed above is addressed in the paragraphs that follow. i Cable Tray Dimensions. Cable tray dimensions contribute to the system mass and stiffness which, in turn, partially determine the system response to seismic loading. Cable tray dimensions refer to cable tray ,7 '

~ m width and depth. The NEMA standard addressing cable tray systems gives standard cable tray widths of 6, 12, 18, 24, 30, and 36 inches.. Cable trays in the data base range from 6 to 24 inches in width. Seabrook trays range from 12 to 24 inches in width. The NEMA-specified inside depth of cable trays ranges from 3 to 6 inches. Cable trays in the experience data base range from 3 to 4 inches deep. Seabrook cable trays have an inside depth of 3-1/16 inches. Cable Tray loadina. Cable tray loading is the primary contributor to the mass of the system. Of secondary importance is the c.ffect of cable loading on system damping. For a nominal 3 inch deep tray, there is roughly a direct correlation between percent fill and weight /ft (i.e., 40% full is roughly 40 pounds per linear foot, plf). Examples of varying data base cable tray loading are shown in the photographs in Figure 4.0-4. Cable trays at Seabrook vary from 20% to 40% full; cable loading never exceeds 40 pif. The data base contains cable trays which are more heavily loaded than 40 pif, thereby enveloping Seabrook for gravity load. In addition, the data base tra3 s range from empty to over 100% full (i.e., from 0 to over 100 plf), thereby enveloping any system damping effects. Number of Tiers. The number of tiers, specifically the number of tiers without transverse bracing, contributes to the overall mass and stiffness of a cable tray system. The number of tiers refers to the number of horizontal spans suspended from a vertical support. In addition to vertical supports, transverse supports are significant to system seismic response. Cable trays at Seabr , ation include transverse bracing at nearly every support. f a ,e trays rarely have transverse supports. Although Seabrook tray 1., .sms include up to 12 tiers, the number of tiers between transverse supports is typically far less than in the data base. Cable Tray Tvoe. Cable tray type contributes to the stiffness of the system. There are three primary types of cable trays described in the NEMA standard on cable tray systems: N.'

g ~ e Ladder Trough e a Solid-bottom A ladder type tray consists of two longitudinal rails connected by individual transverse members (rungs). A trough type tray is a metal structure with a ventilated bottom contained within longitudinal side rails. A solid-bottom type tray consists of a continuous sheet with no openings, contained within longitudinal side rails. Table 4.0-1 lists the data base cable tray types by site. Seabrook cable trays are either ladder or solid-bottom type. Suncort Construction. Cable tray support construction contributes significantly to the system's resistance to seismic loads. Support construction refers to the structure of cable tray supports and the members from which a support is made. The NEMA standard on cable tray systems defines three types of supports: ,x Trapeze e Cantilever bracket a Individual rod suspension e Cable trays supported by individual rod suspension are relatively uncommon, and are not representative of cable tray supports at Seabrook. Trapeze supports consist of two vertical members, typically bolted or welded to the ceiling, connected by a horizontal member, upon which' the cable tray rests. The structural members of data base trapeze supports are constructed from rod, strut, or steel angles. Cantilever bracket supports refer to a large variety of structures in which the tray is cantilevered from a vertical member anchored to a wall, a floor, or a ceiling. Data base cantilever brackets include "L"- supports and "T"-supports (Figure 4.D-5). In addition, at some data base sites, cable trays are supported from cantilevers on floor to-yg ;, .(__,

r ceiling columns (Figure 4.D-5). Data base cantilever bracket supports are especially susceptible to seismic loads because of the dead load moment inherent in their asymmetric design. Table 4.D-1 lists the data base support types by site. Most of the data base sites have trapeze supports or various forms of cantilever bracket = supports. In general, cable tray supports in the data base do not consider seismic loads in their design. Seismic design is not required in the NEMA standard for cable tray construction. Seabrook cable tray supports are of two basic types: Trapeze a Floor-to-ceiling box frame e All Seabrook supports are made of cold-formed steel strut. Each Seabrook trapeze support is transversely braced, making it significantly sturdier than data base trapeze supports. In addition to trapeze supports, many Seabrook cable trays are supported on floor-to-ceiling box frames. These supports consist of two vertical members, supported at both the ceiling and floor, with cross members to which the cable trays are bolted. Seabrook floor-to-ceiling supports are symmetric, which mediates dead load moment. Cable Tray Soan. Cable tray span is a significant contributor to the system stiffness. A span refers to the horizontal distance between vertical supports. Cable trays at Seabrook Station are supported vertically at least every ten feet. In most cases and for most configurations, Seabrook trays are supported every five feet. The spacing of vertical supports of data base trays range from four to ten feet. Most Seabrook cable tray supports include transverse bracing. In many cases, data base trays have no transverse bracing; lateral support is provided only by the geometry of the system (i.e., transverse support of a cable tray run is provided by an intersecting branch run). At some data base sites, transverse bracing forms part of the vertical support. . ' W' F i

Data base sites, normally have no specific provision for longitudinal bracing. Instead, longitudinal load resistance is provided by inherent features of the cable tray system, such as: Geometry (e.g., cable tray intersections with branch runs) a e Vertical supports a Interfaces with the building (e.g., walls) e Electrical cabinet connections e Cable Tray /Succort Connection Details. Cable tray and support connection details contribute significantly to the strength and stiffness of the system. The connection details considered here include: Tray-to-support connection e Tray-to-tray connection a Anchor-point connection (e.g., wall, ceiling, or a floor connections) Cable-to-tray connection e Support internal connections (e.g., connections between a vertical and horizontal trapeze members) A comparison of connection details between Seabrook and data base cable tray supports is shown in Figures 4.D-7 through 4.0-9. Tray-to-support connection refers to the method by which the cable tray is attached to its support. Data base cable trays are typically attached to their supports either with two small screws (i.e., 1/8 inch) or using gravity alone (i.e., in some cases, there is no positive connection between tray and support). Cable trays at Seabrook Station are attached to supports using either internal clips or "Z* clips. U,

i Tray-to-tray connection refers to the attachment between cable trays. At Seabrook Station, tray-to-tray connections are made with eight.3/8 inch bolts. Connection details between cable trays in the data base are typically made using a similar configuration to Seabrook, however data base tray connections have fewer bolts. Anchor-point connections refer to the connection details at the interface of cable tray supports with ceilings, floors, and walls. At Seabrook Station, a special " boot" connection detail has been designed to connect cable tray supports to anchor points. Seabrook cable tray supports are bolted into the boot, which is welded to a steel base plate. At other anchor points, Seabrook supports are welded or bolted to embedded steel channels in the concrete wall. Data base trays have anchorage connection details which are not only weaker than Seabrook's, but which have fewer anchorages per length of tray. Data base cable tray supports are generally bolted to the ceiling with expansion anchors (i.e., 1/2 inch bolts). At two data base sites, ceiling anchorage consists of friction clips attaching the support to overhead wide flange beams. Figures 4.D-7 through 4.D-8 are photographic comparisons of anchor-point connection details at Seabrook and at sites in the experience data base. Cable-to-tray connection refers to the attachment of cables to trays. Typically, cables are individually attached to trays using plastic ties. At Seabrook Station, cables are tied to ladder type trays at every tenth rung (maximum 90 inches) for horizontal trays, and at every fourth rung (maximum 36 inches) for vertical trays. Data base cable trays typically do not have ties for horizontal runs and have ties at every fourth cable tray rung on vertical runs. i Support internal connections refer to the connection details within the cable tray support structure. Examples include the connection of the I vertical and horizontal members of a trapeze, or the connection of diagonal bracing to a support. Cable tray supports at both Seabrook and at data base sites have standard connection details (as specified, for l example, in a Unistrut catalog). In addition to standard connections, Seabrook cable tray supports are strengthened with clip angles and i p I y

triangular gussets at critical locations. Most data base cable tray supports have little or no additional reinforcement. Figure 4.0,-9 shows support internal connection details at Seabrook and at a typical data base site. Additional Cable Trav toadina. In addition to cable trays, conduit is sometimes mounted on cable tray supports. Additional cable tray loading can contribute significantly to the mass of the system. At Seabrook and at data base sites, conduit is occasionally mounted on or cantilevered from cable tray supports. Cable Tray Interfaces. Cable tray interface refers to the means of routing cable between cable trays and electrical cabinets. Cable tray interfaces affect the system seismic response by: Providing a source of reaction to seismic loads a Providing a source of seismic interaction (impact) between m the cabinet and the cable tray At Seabrook Station cable is routed between cable trays and cabinets through conduit or wireways. The conduit / wireways are bolted to the tray, and connected to the cabinet through a flexible coupling designed to accommodate minor differential displacements between tray and cabinet. This type of interface connection is intended to minimize any potential interattion hazards in the following ways: The flexible connection allows relative displacements a between the cable tray and the cabinet, without imposing significant seismic loads on either. The continuous connection provided by the condult/ wireway that routes the cable from tray to cabinet prevents impact between the tray or the cabinet and the conduit / wireway. By comparison, many data base cable trays interface with electrical cabinets by routing a small section of tray directly into the cabinet. This interfacing tray section is often bolted to both the cabinet and the main section of cable tray, imposing seismic reaction loads from the t

cable tray system onto the cabinet. Alternately, some data base cable trays interface with cabinets by abutting the cabinet, or conduit attached to the cabinet, without positive connection. This creates the potential for pounding between the cable tray and cabinet structures. Interaction between cable trays and electrical cabinets has never caused damage in past earthquakes, in spite of the general lack of design provisions to accommodate seismic interaction. Cable Tray Layout. The effect on seismic response of cable tray layout can be illustrated using experience data. Multi-directional cable tray systems are difficult to analyze accurately or to mount on a shake tabl e. Cable tray systems in the experience data base are comparable to Seabrook Station's cable tray system for cable tray layout. Cable tray layout is a general parameter that includes the following components: The extent of cable trays within the building (i.e. the e typical length and directions of cable tray runs) The relative configuration of intersecting sections of the e cable tray system The extent of cable tray runs affects the stiffness of the cable tray structural system, which in turn affects the response frequencies (within the range of seismic excitation), and the mode shapes of the tray system. A secondary effect of cable tray extent relates to the seismic input imparted to continuous cable runs by different sections of the building. Short runs of cable tray typically receive a uniform seismic input from local sections of the building. Extended runs of cable trays receive seismic inputs that vary in amplitude and phase, according to the seismic response of different sections of the building. The relative configurations of intersecting sections of the cable tray system affect the stiffness of the cable tray system, and the seismic reaction loads imposed by one cable tray section on another. Cable tray configuration parameters include: 1 l m Spacing of intersections 1

Angles of intersection e Relative mass and stiffness of intersecting sections e Details of attachment at intersections a This in turn affects the response frequencies and mode shapes of the tray system. The parametric components of cable tray extent and configuration might be generalized as the complexity of cable tray layout. The experience data base offers complexity in the layout of typical cable tray systems that is comparable to the complexity of layout at Seabrook Station. In other words, data base cable tray systems are typically of comparable (or greater) extent than the Seabrook systems. Data base cable trays include tray sections intersecting from a variety of directions and angles. Data base systems include atypical cable tray details such as offsets in cable tray run, which create potential weaknesses in longitudinal load resistance. Data base cable tray systems include a variety of interfaces with electrical cabinets, conduit systems, structural supports, building walls, and floors. The complexity of cable tray layout also includes the potential for interaction with adjacent fixtures. Data base cable tray systems are frequently routed in congested areas that include other fixtures such as piping, conduit, catwalks, or structural steel. Since the support of data base fixtures is flexible compared to typical nuclear plant i installations, there exists the potential for substantial sway and seismic interaction between cable trays and adjacent fixtures. In spite of the high seismic interaction potential in data base facilities, there are no instances of interaction damage to cable trays in past earthquakes. The various components included in cable tray layout, as well as typical examples of the complexity of data base cable tray systems, are illustrated in Figures 4.0-10 through 4.D-12. y, .V,

Cable Tray Locatioq. Cable tray location within a building structure affects that level of motion experienced by the system. Of particular interest is the elevation of the cable tray system above grade elevation. The amplification of seismic ground motion generally increases within a building with height above grade. The location of cable trays at the Seabrook Station ranges in elevation from 47 feet below grade, to 60 feet above grade. The building elevation of data base cable tray systems ranges from basement locations to locations in steel boiler structures, over 100 feet above grade. TvDe of Buildino. The type of building affects the amplification and filtering of seismic ground motion into the supports of a cable tray system. The parameter of building type also ir.cludes soil conditions at the site (i.e., rock, deep alluvium, etc). The various types of structures found at the power stations ar.d industrial facilities surveyed in compiling the experience data base offer a wide diversity of building size, and flexibility. This in turn suggests a wide diversity in the amplification, distortior, and filtration of the ground motion experienced at the various sites. Data base buildings that house cable trays range from flexible structures to structures comparable in stiffness to the buildings at Seabrook Station. l One extreme is the tall, open steel-frame boiler support structures typical of fossil power plants. Steel-frame boiler structures are typically five or more stories high, and contain the massive furnace-boiler system, usually supported as a pendulum from the top of the l structure. The boiler system is usually free to swing within the steel-frame structure. As an example of the flexibility of boiler structures, a response motion record taken near the top of the 169 foot tall open l steel-framed boiler structure of the las Ventanas Power Plant recorded a l primary response frequency of approximately 1 Hz during the March 1985 Chile earthquake. The record measured a peak acceleration of 0.80g. with a duration of strong motion of about 60 seconds. The boiler

structure contains a system of cable trays which was undamaged in the earthquake. The bulk of the cable tray systems included in the data base are contained in two-to three-story steel frame or concrete shear wall buildings, such as the turbine buildings of power plants. This type of structure is generally similar to the Seabrook Station structures (other than the reactor containment). Typical fundamental response frequencies for this type of building range from 1 to 5 Hz, which corresponds to the frequency range of maximum energy content for most earthquake ground motion. Table 4.0-2 summarizes the types of buildings and the site soil conditions for various data base sites. The Seabrook Station building structures are somewhat stiffer in comparison. Based on building response analyses, fundamental frequencies for the Control Building, Primary Auxiliary Building, and the Containment range from 5 to 10 Hz. This frequency range is slightly above the range of maximum energy content for typical seismic ground motion. Seismic Ground Motion. The level of anticipated seismic ground motion forms the basis for the seismic design of cable trays. Seismic ground motion is defined in terms of three components: Peak ground acceleration a Duration of strong motion (typically defined as > 0.10g) a Frequency content of ground motion e These three components are characterized either directly or indirectly by a ground motion response spectrum. The basis for the seismic design of cable trays at Seabrook Station is represented by the ground motion spectra of USilRC Regulatory Guide 1.60, normalized to a 0.25 g peak ground acceleration. This response spectrum is compared to a range of data base site response spectra from various earthquakes in Figure 4.0-14 (plotted with 5% damping). As shown in the figure, most 0

data base ground response spectra are either comparable to or in excess of the Seabrook design-basis spectrum. 4.0-6 Conclusions The comparison of cable tray systems at Seabrook Station with cable trays in the seismic experience data base has demonstrated the following points: All parameters associated with the seismic capacity of s Seabrook cable trays are enveloped by data base cable trays, i.e., data base table trays are similar or weaker in all aspects related to seismic capacity, compared to Seabrook cable trays. Most data base cable tray systems have experienced seismic s loads comparable to or greater than the seismic design-basis loads for Seabrook cable tray. systems. Cable tray systems constructed according to normal industry s practice have more than sufficient margin to absorb the seismic loads anticipated from moderate earthquakes (such as the Seabrook seismic design basis), even without specific seismic design provisions. The capacity of cable trays to survive seismic loads is not sensitive to details of cable tray construction or layout. By a comparison with cable tray systems that have survived past strong motion earthquakes, it is apparent that the cable tray systems at Seabrook Station have more than adequate capacity to survive their design-basis safe shutdown earthquake. I m. -ej

Table 4.0-1 CABLE TRAYS IN THE SEISMIC EXPERIENCE DATA BASE PEAK GROUND

CABLE TRAY SUPPORT EARTHOUAKE SITE ACCELERATION TYPE TYPE CONFIGURATION

, San Fernando Sylma r 0.50g Ladder Unistrut Trapeze Earthquake Converter & Trough 1971 Station Valley Steam 0.309 Trough Unistrut Trapeze & Plant Cantilever Burbank 0.32g Trough Rod Trapeze Power Plant Glendale 0.279 Trough Rod Trapeze Power Plant Pasadena 0.18g Solid-Steel Floor-to-Power Plant Bottom Angle Ceiling Cantilever Saugus 0.359 Ladder

Inistrut Floor-to-Substation Ceiling Cantilever Point Mugu Ormond Beach 0.20g Trough Unistrut Trapeze Earthquake Power Plant 1973 Ferndale/

Humboldt 0.30g Trough Unistrut Trapeze Humboldt Bay Power 0.25g Earthquakes Plant 1975/1980 Imperial El Centro 0.429 Ladder Unistrqt Trapeze Valley Steam Plant & Trough Earthquake 1979 Drop IV 0.309 Solid-Steel Trapeze Hydro. Plant Bottom Angle I Coalinga Union Oil 0.60g ladder Rod Trapeze Earthouake Butane Plant 1983 i

Average of Two Horizontal Components of Ground Motion I:'

'M V:n 3.ki! ',

Table 4.0-1 (continued) PEAK GROUND

CABLE TRAY SUPPORT EARTHOUAKE SITE ACCELERATION TYPE TYPE CONFIGURATION Morgan Hill United Tech.

0.50g Ladder Unistrut Trapeze & Earthquake Chem. Plant Floor-to- . 1984 Ceiling Cantilever

Santiago, Renca 0.35g Ladder Rod Trapeze Chile Power Plant Earthquake 1985 Rapel 0.31g Ladder Steel Floor-to-Hydro. Plant Angle Ceiling Cantilever Laguna Verde 0.30g Ladder Steel Trapeze Power Plant

& Trough Angle Las Contes 0.259 Ladder Rod Trapeze Hospital Las Ventanas 0.30g Ladder Unistrut Trapeze & Copper Refine. Cantilever Las Ventanas 0.18g Ladder Unistrut Trapeze & Power Plant Rod Cantilever Mexico Infiennillo 0.15-0.20g Trough Unistrut Floor-to-Earthquake Dam 1985 Ceiling Cantilever La Villita 0.159 Ladder Unistrut Trapeze Power Plant Cantilever SICARTSA 0.15g Ladder Steel Mounted on Steel Mill Pedestals Pipe Gallery

Average of Two Horizontal Components of Ground Motion i

5, 73

Table 4.0-2 BUILDING / SOIL TYPES ESTIMATED EARTHOUAKE SITE PGA** BUILDING TYPE S0ll TYPE Seabrook 0.259 Reinforced concrete Rock. Station shear wall structures San Fernando Sylmar 0.50g 2-story steel Sand, silt & Earthquake Converter frame structures with clay. 50 ft. to 1971 Station penthouses and

bedrock, reinforced concrete structure.

Valley Steam 0.30g 2-and 8-story braced Deep alluvium Plant frame structures; to 500 ft. first 2 stories with reinforced frame and shear walls. Burbank 0.329 Five-story braced Brown, sandy loam Power Plant steel frame structures; to 25 ft., dense first story with rein-sand below. forced concrete frame and shear walls. Glendale 0.279 4-stos. .rtially intermediate Power Plant braced steel frame alluvium. structures; with rein-forced concrete first floor and basement with 2 lower stories. Pasadena 0.18g 4-story steel braced Intermediate Power Plant frame structure with alluvium. concrete walls. Saugus 0.359 l-story reinforced Alluvium. Substation concrete structure. Point Mugu Ormond Beach 0.209

2. story steel frame Alluvium.

Earthquake Power Plant structure with rein-1973 forced concrete walls. i

Design Basis Earthquake
- Average of Two Horizontal Components of Ground Motion i

( 1 a

?

Table 4.0-2 (continued) ESTIMATED EARTH 0VAKE SITE PGA** BUILDING TYPE SOIL TYPE Ferndale/ Humboldt 0.30g 2-and 5-story steel Deep alluvium Humboldt Bay Power 0.259 frame structures and to 400 ft. Earthquakes Plant 2-story reinforced f975/1980 concrete frame structures. Imperial El Centro 0.42g 2-to 6-story steel Deep alluvium. Valley Steam Plant frame high bay and Earthquake braced steel frame 1979 structures. Drop IV 0.30g High bay reinforced Rock. Hydro. Plant concrete shear wall structure with partial reinforced concrete basement Coalinga Union Oil 0.60g Cable trays are Shallow alluvium. Earthquake Butane Plant supported on steel 1983 racks running through the facility yard. Morgan Hill United Tech. 0.50g 1-story tilt-up Shallow alluvium. Earthquake Chem. Plant concrete structure. 1984

Santiago, Renca 0.35g 3-story braced Deep alluvium.

Chile Power Plant steel frame. Earthquake 1985 Rapel 0.319 High bay reinforced Marine sediments. Hydro. Plant concrete structure with a 4-story rein-forced concrete frame mezzanine, adjacent 'o concrete dam wall. Laguna Verde 0.30g 4-story high bay steel Rock. Power Plant frame structure, part-ially braced, with re-inforced concrete shear wall first story

- Average of Two Horizontal Components of Ground Motion

Table 4.0-2 (continued) ESTIMATED EARTH 0VAKE SITE PGA** BUILDING TYPE S0ll TYPE Las Ventanas 0.309 High bay steel frame Compact fluvial Copper Refine. structures with sand to 165 ft. bracing Las Ventanas 0.30g 5-story braced Compact fluvial Power Plant steel frame structure. sand to 165 ft. Boiler structure is a 169 ft. steel frame. Mexico Infiernillo 0.15-0.20g High bay reinforced Rock. Earthquake Dam concrete frame and shear wall structure, with 2-story steel frame mezzanine, com-pletely underground. 1985 La Villita 0.15g High bay steel frame Rock. Power Plant structure with 5-story steel frame mezzanine.

- Average of Two Horizontal Components of Ground Motion

9 t oo. i so I-t so - i ao s. s,ime, c....,ier sionion ...... usmac noe ovwe i 80 HVAC, _{ ' 20 ROOM N CONTROL l o ROOM g L 30' i oo $ o a3 sWITCHGEAR ROOM o so POa= 0 See N g [d d L'! N ~ o so l o ss N CABLE tf SPREADING L = Rxm ' ", o is' o ao o as'o ao'o 35 o 40'o s, io e Freover cy (H2) kMh'$7h-II I:hj ! c r 7 f I I.-- 48' }.- 2 4 **+{ l 24N al A r 2. H 24"H ' X I E-% i,' N am f. P' , r 7 12 ** I 8 2' W 24**M t kb J L f l g,t cj i .a t NCI E'><l J x 24-P n a l F l P y 7 D---C L g , r-uWO4 4 W24% L J Q figure 4.D-1: The Sylmar Converter Station, affected by the 1971 San Fernando Earthquake. Average horizontal peak ground acceleration is estimated at 0.509 The response spectrum is represented by the nearest ground motion record taken at Pacioma Dam. The cable trays are located in the basement and suspended from the ceiling of the second floor of a 2-story steel frame building. Typical cable tray {jTg> supports are Unistrut frames in trapeze configurations. Some supports are framed '= ' a - directly into adjacent concrete walls.

2 00 ' 40 Unumy '60 N \\ s.0 L t 20 ... - - U$NRC R.g Gwsd.160 i s 00 rT1-m lam n 5 [ F5 ,/ 0 60 <. I l CONTROL e roa-m e l '- eaa.os,e

- x

%..a ~ ',Q:afa L .4. ,,...ac, m a m r-su n T w O E Ita i AL ) OT I ~ am p n c 36" 5* 18"Tra ys m ~ + w- -==:: c:e gp w ra w w-l Q9 _J_3 4 s2a Figure 4.0-2: The Valley Steam Plant, affected by the 1971 San Fernando Earthquake. Average horizontal peak ground acceleration is estimated at 0.30. The response spectrum is represented by the nearest ground motion record taken at 9 8224 Orion Blvd. The cable trays are suspended from the ceilings of the first and -t ' h; >1j second floors of an 8-story braced steel frame structure. The cable tray supports consist of 'ight gauge steel members in framed and braced trapeze configurations.

s.4

s oo e so i so i so e g

20 f

tic.aine si.. ,.ae / --- - - usanc nes co.a. i so

o.-

v v . - x-

s. m.o..

" N _ -- - / \\ ,o.. o.,, l 0 eo g.p. y ._a a to f ' ' ' ' ' - 10 " " * * *J... e 0 oo h oo So io o is o to o si o ao o as o .o o ~ ') e,.av.ac, ins) l STEEL BEAU f ~d JL JL t 1 4 % 4,- 6 l r r a-y az-a r=__m ,2- ,,t. g I [__ 3*" we-o r =_-a m m r AM r -~ 9-be t 2d M h 1:wn I \\ H 12 H P 1000 UNI $1ftUT Figure 4.0-3 The El Centro Steam Plant, affected by the 1979 Imperial Valley Earthquake. Average horizontal peak ground acceleration (measured at the site) is l 0.429 The cable trays are suspended from the ceilings of the first and second floors of 2-story steel frame buildings adjacent to the turbine building and boiler structure. The cable tray supports are Unistrut frame trapeze } configurations. l l i L

\\ \\ l i -I ,s. l

,.., - )-

l_ In:. m - , / ?y +..X ^ ': ".: U.,-h.[ki ME h k-.

namu w;n 1

.,'./* l g f g: t % f$,ggi,ged * ., - e,,' s iss a ;

w,

,m. ; .lQ[_- y ri@ : - l

1-q; p

f. +.. ' j h 'g '. A =. .e d': ~ y ,s

~

r b Y... 6 ? , Q.. s j<,lf[ '{g N. $- ~.4,4 5 ' I- ' s.f r :; A' I g.

7 '.

4- , 5x jl. S,f. N? R- '~ sa. j : - ( pr. L ',;g .w.J, w.,2, u n a n +- L .,.. J??f lf- ? .? i f~ f, kE ? %.y?fR g.v x. [. - h +-: -.g

t..u,wn

-Q 'j__': /', j, " Q Y '. /,,pes.- y ~ g b.?, .g ,.,}. 2.. Fm o J * ~~ l, _. ? 2,= = t e f.:.-; %,' y ;.g g _ b ~ '.- {7.. ' <.g. g '.,,: y , + 3, .~.M ~ % ; .y.. . r. .s - t n v.,. -t w,c., l +$g,.p.%Z A. w"t W. t "g,n' '4 , ?. /. '%R

Rj ;ft;Jk,a,,,Q

.s. :s. r

;fhi

-t t.,..,e. A..t. v.a%: ;g,;t:; p c.; us. Y M7:- e,?; f6-.- ..-t ..c. ff,,.s,- ,J -" y :. : :- *... w, p -.W.5.3 s' .s ,e r Figure 4.0-4: Da'.a base cabla trays range from almost empty to over 10C% full, as shown by El Centro Steam Plant (upper photo) and la Villita Hydroeler.tric Plant (lower photo).

,s-- .?* ,yJ9 ' &..~ ene d # ( ~.. y::*.-{- ...Q , ;..;.y % ; .,,s y?:+# - a .... ' v. s,~ t? 7 W '. L.. : ,g".A - ' T. 4 - r T'. ' ; l. N \\' l ' h -- . t(.,. c.{ A" W @,'ls,. l p'.'Ch')l}y ry.f_ l 7. f hY Ti: QAA ' l(_ F ;.: ~l[Q Llq~} )[ _fY_ $7( ' ' g g 88 th, 4, . g %s. ' L. DA... rsmir# ,. s f l l f l l, - a,.r').Y$k v-.. v; e .= es q l, r_p. ' N2 - a - _s. Q," y. + [ ) *i,.- _ *q + t. y %,' -3 u ..a..,.. .. t 4,. ~ ._l.:'

r. g+ -

'. '; ',j.' *:-. l l . _ :g - 5 g, .,., 7...w 3 i. L'"yg (Q kI p -L W j,s ? h. t I, i r 8 b 4 f ~~- ~ '[ y .-;..g 3 ! t. . }] :jg .L[ y*f' 'y i-l;n.h.-;2 4:; ,.g g . % t . sp o s ,p c!.$.g- .. n 1 i**.. - h i;. i- 'T' "4-L [ ', u '}% '}- ~ l%EY_.?i[. ). f. l ?/ _h.' O h ;'J ?F ' ' : "l : L... Figure 4.D-5: The experience data bas +) cont ains exa'npler of both T-configuration (Valley Steani Plant-upper photo) and l.-configuration (Rapel Hydroclectric Plant-lower photo) cantilever bracket, cable t ray supports.

l l l l .... ',e,r%R..;).. Av .~ [s. -, ~ +. i.' 2 %. + ;X4. " 4 s., g ( y'[.c - Y,N" ~ N g' '... + - 'fj* =,., 4 _, . h; l lS,, .9 l-M. s .et ~

g

.,..3 .ll y i .g .....w. k..y. - l m~ 'cuyl.$h.L y.l. 'Tfl.. - _[, e ....v 3 '3..-

1.,.....

. 3.4.,c.,. s.,,g..g. . ' ' -.. 4.g;,su,,.. .. l y,... , - -. -~ *

3 E -

6 ~

.. r, r.m.. -

1.

9
p,,.

c -8L i ' '+ ~

.....,.. ;...,. :..."... w%,3 *

- # c,.. w -... n.. i .r 1 .[ .-~,.n..- i y 4 s -q _ t 0 =s.-, a '. ' 1tre f.... . - {. - - y ;i ' ,r. .w g.--,'.))... i I -.' +.. ~ x, k.. 3 y+; i r* .j.

p..

', - [ " r '.[ ;]() 4

.- w i

%..h.9 d.5. f. 's g.- l ' ' 3.., t l 3.%w,.C.;f, iy.g: ,{,; > l ":.7 f*, .. E ' 0_.R;_-(ltp g*? Y E A V4.q.ptj l :" L.g + x. 3 r ;.v,,.;. . v.i y,65 Figure 4.D-6: Typical floor-to-ceiling, cantilevered cable tray sup;> orts at the Saugus Substation (upper photo) and El Infiernillo Dam (lower photo).

1 l d i ) I[' ' ..g '

4. / < gg - ' - -
. l 4.,

9 x. t ; b.-\\g. 14~ ~.. -L,,.' gf,y '.L;.

m. c.h,.,

., c. s.t. , ~ f,~:;t w. M:q:, k c r- ,v;..,... ;.n#. _. 7 -.. w .v.. 4.e z.., 3

f.,p.gg ;.

y.a e.c,.. ;4. ;s :. >en..

y - :, z.p.,. x,- 5 j .. p. m..>. s.;.:.pA 'p::.2 a - :. c.... .y...: hy....y..,.; y. .y ' ; ..%s.W; L. q."".,'.,

g$. ~ ~.T' 3 $

w....,4, ,..o.. e, ,,f, .:. 3.g : s. a,.: f - 3,. .g y ~! ' ' ' + J .g. .v ' yll -..: s -l., :. :. y... -9 :, L '; ~.. 1.g..M,.w;e y;; - g:. 42 i i;,'. -. :35 7.;- , [- g

y. :

n.. Q.:..,.

q.

49.

i c ......e, t,.JJf. -m.,3 '.- ~ f g!,fQx;,.. n. g%g.

e...

.. w..

7.,3gt;7-?y;.? y.%.6,.w.,

ay :m%. .e g.ey ~. ;, v. e s v. q,.:< e.. 8,?- .u - a

m. (6...s

, e t. 3;; a,. x ~ 7. ?L. ;, :f:.,. '. - j).,,. ;.g.g,y.$.J.g$k;l c a:J. h. . l: ..[D, '.h .iME .N 7_ -p:y. Q...,3, Q. qsdf y;. 4 n._ p; g, j. . c f =;?,g; w;n L :,.' g; 1.~ .. :... 4, . p.e p, :'p., .q . -. r. ..v/...,,.. y' :4g..q,y. .-. g,. ..n.. - 9..., , e. '. . j.., : :..q, g),..p.. - g>.. e.t,gye;. . t,.1 f s , 4,g. p. i .. 3 4.:.,..... 7 s. - - ~ ... - L. . v y, -m..y#, m.;y m,. %r. v-;., g. . &+.m. w s.. w u c..r ;.-... - 1 \\ l i f ', jg ~ f'g,, g y 3 s %..($$.1 m.qg..ge\\],gx z 4.. ,3.;; - y q,,,.

.p.

- 4p. i (, n. x -4 -he e, - 7.; - 4_, :.,R ,.E.yR : 1 ,e w g s. .,y,.,...,..., *... y,, .9 t i ,* q.:4 x l ~,s - " A1, .y -~ y.

~. ;, ; {; -,,.. +._ g 3 ;g y s.3.g,. e: k.:.,-

L a..... n,.;.

1-p.,

%.x. g. l [yyy;: fh ,I.; L*t . g.. q :. c ..[ '.. l n ). &; '. W :sn.v e,p.% 4: J c y n.. g q, q y,.. y*... t 7 '.;'e_ g n +37-s y \\

a..

N ;.5W - I': e.%.y v y- .x }. Wv$,;, f .y en~. 9 .., -., r. t i 7 C ;.. .t..& .h % t i i Figure 4.0-7: At data base sites, such as the Drop IV Hydroelectric j l Plant (upper photo), cable tray supports are typically anchored to concrete ceilings using expansion anchors. At Seabrook Station, supports are typically anchored through a " boot" connection or bolted to embedded steel channel in the concrete ceiling (lower photo). i f .. c _._.._,_,-,m,.

l l l i .A g[ 2;, g 4 4 '? ' ' ? [}M..,' 1 >a = t. -p g i \\ '.r y. 4,, ,f ( i-i

b. < '

i i Ig gg3 0- ) 4 cqp f

l8

,, ;, o y. 3 1 I I - 4 / \\ l

A.

} / %4 ..s ..i:. ( .54-

l I

tg; i l i I Figure 4.0-8: At data base sites, such as El Centro Steam Plant (upper j photo), cable tray supports are typically attached to overhead wide-i flange beams using friction clips. At Seabrook Station, supports are typically anchored to overhead beams by welded connections or with a " boot" connection (lower photo). l f f i 1 l l l 1 i

s 's \\ \\ N \\ g \\ s N N s \\ a \\ N N N \\ s t [ s 9 l Figure 4.0-9: Typical support internal connections at the Sylnar Converter Station (upper photo) and at Seabrook Station (lower photo). l i l 1

i i t? y. Ij I l l l j s t l l i 1 I \\ ./ ) '**I, .r I l l l l i'. f;,..};

Y i

t a,.;,:. i d#$1{I, m :i-l _.;t y., k I i - : m. - -=..- Figure 4.0-10: The extent of continuous runs of cable trays affects the stiffness of the system, and subsequently its response frequencies and mode shapes. Examples of extensive runs include the pole mounted cable trays routed for over 300 feet through the power house access tunnel at the Infiernillo Hydroelectric Plant, (upper photo). More extreme examples are found at the neighboring SICARTSA Steel Hill (lower photo), where continuous runs of cable trays and piping, supported on racks i J extend up to a mile betweeri different sections of the plant. l l m,-:. <p f_ i !L----.

. _. _.= _ ~_. l l l = l l l . ::: :; c..e. ~ L. g ;- g y . x ~y ' - - ' + ~ '.. ' ( - ,,g T*%.s, %.?jj;.Q g..5 W 1 c * :**.. :. g'~2fs t. -d n f4 .i &'h ,.l: *jj?$$,R. Y:~'#0'.Nhb '2:'.$$' '. N^ '3 h.Yr. ; l .f, ;,'. &Y&h& $.;$.W;W WQMNnll?.* Ql~GM.4: pef.5 l l kf h,[ - .h i d

.Q.. ] Y.5'.i!.:'

VQ.'.hi'1 S,.* b ^:.'.r -:.- L. M e'p':* hy...g, ~

;;k;-Q: ~,- <l?g.'.f,:,'b $ %.

. :<Sp' %. n 1 J E w a *. t..%.;,.-p.y,.%.;Z.i '".' Jd. .;y, g..t'f gc.;.;gyi.h,.% W_.:<, $q.:. 3h,....w ,;' %a' )/7s ',p{;4 p%,q:. k;',:~;7.f[(ri:.g% . A b. ~3~ h 9'+, y 6> v :,.y;{I4 ., ( +.e,. ~ $. h,::s /;.. ,5 7.$:./ :.g. g...a<QM.Y .4Y...- 6 e / 4MS; 6 L rim 4; G '

W

.P. .r_.:;:y..{p : ) !f.e,;:.'._:*QQXl,*y_.,lI[h s 4$.'y;f.f. ~C~-

yp *

~~< N^ ~jf ';; *R. .,:;i..}f,M*~,..;,! .. v;; l D ' '%.?:, L:.Q% t i%

( 4 k

i ?(. h? N.... ?% ~f - ~ t /;;p.Q,.%..%. y. *. %....ph.j5 :. 'G.% s,+..'. t.

..u.~.,.....

y + + r m..., s l %.., y > ;; y., a. y.p.. .y ...e .. - o. ' ? : - ~. e y.t.w:2 M.;:.h V.~-. s.n..s. y;. g.... .. ;. ?

4%

s. e . s. '::;.. yd@'R+...Q.. [:;g$h.

*,'y&,*'

Ry*h N. t.%a}:

'N. t. .t .. W6.. :.y,. .r i

Q{

-:. t : :.. s. N.t.::. f:l?. i.Q*y.! ~'X,' . >. ;.n... 4. . N' c, Ns 4 l y. ~.a. a y a s M k RLM V. 2.*.S M f r n i j 1 L 1 4 4 qs l t l i 4 a. i I a + I l rigure 4.0 11: The relative configuration of adjoining sections er I cable trays affects the stiffness, and subsequently the response l l frequencies and rnode shapes of ttie system. The angles of intersectiof, i of adjoining trays and the details of their connections affect the i magnitude and direction of the loads icposed by one tray sectior, on i another. Examples of diversity in adjoining tray intersections are j ,1 shown at La: Ventanas Copper Refinery (upper photo), and the El Centro Steam plant (lower photo). I 4 1. i

t 4 j ,z. l ~ i N' i h 1 l 1 i k I 1 ~ i .c 1 r ( ~_. I. i i, ' +A ~ . mes # . p,g ..\\..#~ ~ \\ 's \\ 1 m my \\ b, ,sP., \\ 'kd',- W.,. _ ,ke / ' ..s r r 1,llS 'i, y stu e. m.8,s"w, i i l 1 Figure 4.D-12: The complexity of typical data base cable tray l configurations is illustrated at the El Centro Steam Plant (upper photo), and the Sylmar Converter Station (lower photo). l l l i I l l 1 h l l

1 l _.h i0 'Nj * \\ .Qge: . l($ vi X. y f. _y-5. - - - .$.,, M [l J, '. l[f.,., _ *_ F, ]

j. -

. [, I\\l kmb. s s._ w-s 1 m-d.. .y' '..: ,3

t

r. t.. i

f5 7p5. _- _.);.

-:.=,,(,., ,37 sf-3 +- - , p: ,s t W, : - .?.., \\ l j.. 7 .? e; ,n f. e _ V.y. - l ~ ; [. b h u T; f ) g f, ) l i .s ,. 1., e-e : if l.j, }Raint,g (... 3 ' ,.N e ~ .... _, +< g e

3..f c f ~ ' '.

4- - j - ^ ~ g: 9 t.. et);., g o .=<;1 - 3 I .: r 4 ,..a ..g' g s. .3

?. $'

). jy _

Y 1

,... e; n, l f;[95.f.l f[.i ;q }. I il. 1 N', pay [:gy._ [.f M. ' d -( .)_ y;= u,y. x.;. l '-Qh, g y

f v

w.. w. .p. y%.l;Qj ' ~,[_{ [ {$ g' f, ~ ^ gy 9y.; y c.. '~ h. :. w.

x..

.s r w l,;y

^

't I ,V- .;I 4 'll*:fx. g i' n g J\\% / @??fh, = ; f [ * ^g.j n ~ #, . q j f m;w. \\w,"p,.,g,/. n, :'. >, - e.,,.,.. f'j. s;,. - ^ ns g,: ww. b ,,, w. _ yg,. " ; _ 7.,.y ' '3- 'f ..t t. g.-

,.?

3,.n a. n l -~~M'i t'.ld% t L. -.. ' %. % l ? ~ s. \\ ~ _lhffi.h Y ? 7 y ..,..q -> n ;.. t g 4 W u t u- _p .: Cy ' - p;~ t *,,'. '.; '. m +~ % ;,;p fyy ~r y I .cgpg g. . ].' .h --( s}. V)

f Q..-

V ? 3,. :.; q.n{.,'y' a%y.. 2 q y% ' . %,,.,.Y. f r _ t. p l_..; 9' l l figure 4 0-13: Data base cable tray configurations typically offer the i pctential for seismic interacticn (impact) with adjacent fixtures, such as p; ping, at Las Ventanas Copoer Refinery (upper photo), and conduit, at th: Valley Steam Plar.t (lower pFoto). Seismic interaction has never becn a source of damage to cable trays in spite of dense configurat ions and flex'ble support s of pipir.g, conduit, duc t s, and cable trays in data base facilit.es. a i

.n.i - =:..;-...-,a. t 2.00 - i g. T-1.8 0 - - 1.6 0 - - ./I Centro Steam Plant ylmer Converter Station 1.4 0 - - brook Station SSE Valley Steam Plant 1.2 0.. urbank Power Plant i g Humboldt Bay Power Plant -l v I.a Villita Hydroelectric Plant ,o_ 1.00 - @3 0.80 - j ,l s oo y ~. 0.60 - ~ O.4 0 - - - '~: 2, i 0.20 -]! ~ ~, 0.00 l l. l 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 Frequency (Hz) H gure 4.D-14: Range of data base seismic ground motion response spectra with PGA 0.25g). superimposed upon the Seabrook Station SSE (USNRC Regulatory Guide 1.60 i ~= a m .}}

ML20141B832

Contents

Text

SUMMARY

7.0 REFERENCES

+

:

7.0 REFERENCES

Navigation menu

ML20141B832

Text

SUMMARY

7.0 REFERENCES

+

:

7.0 REFERENCES

Navigation menu

Search