ML19225C872

Responds to IE Bulletin 79-02,Revision 1 Re Seismic Analysis of Concrete Pipe Support Base Plate Design.Forwards Rept Statistical Sampling Program to Test Concrete Expansion Bolts, by Bechtel Corp

ML19225C872
Person / Time
Site:	Arkansas Nuclear
Issue date:	07/06/1979
From:	Trimble D ARKANSAS POWER & LIGHT CO.
To:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
Shared Package
ML19225C873	List: ML19225C872 ML19225C876
References
1-079-1, 1-79-1, NUDOCS 7908030111
Download: ML19225C872 (11)

Text

--

,n t f >g$ 1 s.

. .n J L

~

1 i!' ,

Chib ARK ANS AS POWEA & LlGHT COivPAN'r' '

-',-.]i

m. /noc c v A O.)... $. < Lc mut A *r- *e ,., 50 e r c. A - - ,

!Ov i--.I Je . -

u ,1, July 6, 1979 1-079 1 w,_nla_i.

J

?.'r . . . V . Seyfrit. Directcr Of fice o f Inspection L Enforcemnt Regica IV U. S. !bclear Reculatorv Grrnission 61 ! R r. '? '.1:a D-ive , Suite 1000 Ariington, .exas .,, c, u, i J Subjec;- Arkansas Nuclear One-Uni ts 1 & 2 ,,,30-313 & 30-200 Doc:. et . ..os.

Li ce :s e Sos. DER-51 & .'GF-o IE Eulletin 79-02 eile: 1310.1, c,-la!0.1)

Gen:1ern:

'Ihe follo:<ing is provided in response to your ILE Bulle t in 79-02, Rev. 1.

Units 1 and 2 are individually discussed within each response.

u Im..

Verify that pipe support base plate flexibili ty was accounted for in the calculatica of anchcr bolt loads. In lieu of sunporting ana' vs i s j us t i: i ng the as str: t ion o f ri .;:di ty , the base p;ates shculd be ccnsidered flexible if the u.st.rtened distance het.seen the atrt+;r selel to *he pla* e a"d the ec;;e a: tne base pla*e is grea*.er *han *% ice the thicer.ess Of the pl2te. Ic53 conservltive ac c ep t 2".2 2 criteria.m

• be j us + - ' " ' -- d t' .e ;us*.:ication suu.it*ed as par

• of the respctse to the il.;:i.. If the base p;1*e i s d .: t eri n-xi t o be

• l exibl e , then reca.cula*e the bel

• l o ads us i ig r. appr cp r i l t e anal ys i s . If poss i's.e this is to ce done prict to *estin' aof anchor hel t s. ~hese calculated bo!! l o 1d3 are referred to hereinaf ter as the bolt desi.~n loads.

A desc ri,- tion of the ana'. ft ical t.mdel us ed to .eri:, t' rat pipe s=por t base plate flexibility is ac:cunted for in the calcuia-t ica o f ancho r he l t lo adr i s to be efec.i t t ed wi th your respense to the Bulletin.

4g 0 o,6 79 08030 W

. .e - s. ==m smm, s s . s s .,

e, --- e

1-079-1 2-079-1 Yr . K . V. Sey f r i t July 6, 1979 FISECf:SE All pipe anchors (seisnic categorf I) and pipe supports (seisnic category I) for greater than 2 inch dia~eter sys t ers were reviewed to dete=ine i f they crployed base plates that were secured wi th ccncrete expansien anchor bolts. Identified base plates were then re-viexed to de temine i f they were flexible using the above cri terion.

All flexible (as defined above), pipe anchor and support base plates using concrete expansien anchor bolts (C3's) are being (re) an-alyzed to account for plate flexibi1ity, bol: s t i f iness , shear

-ter_sien interaction, rnini:n n edge distance and prcper bol t spacing.

Depending upon the ccnplexity e f the indi . idual base plate con-figuration, one of the following ~e thods of analysis is being used to deterrnine the bolt forces (bolt design loads):

(i) A ccr:puter progran (ECLTS'), developed by Eechtel, is used for base plates with eight bolts or less. The revie.v of our base plates revealed that the nnj ori ty of then were anchored ei ther by 4, 6, or 8 bolts. The plate thic' cess uns usually between i' to 2" and the large plates are not generally stif fened. For these types of base plates, an analytical fomulation ('BJLTS") was developed 0.hich treats the plates as a bean on nultiple spring supports subjected to rnrents and forces in three orthogonal directions. Based on an-alyt ical considerations as well as the results of a nu+er of re-presentative finite ele:ent analyses of base plates (using the "ANSYS' code), certain em irical facters were introduced in the sicpli fied bean ~rdel to account for (a) the effeet of concrete founda t icn (b) t he two way ac t ion o f lo ad t rans f e r i n a p l a t e.

These f actors essentially provided a way for i .troducing the in-teraction effect of such paranetric variables as pla te di ensions, a t t ach.~er t sizes, helt spacings and stiffnesses on the distribution of ext e:~;al lo ad.s to the Dolts. 7ne resu l t s a i a n:rbe r o f Cas e studies indicated excellent correlaticn be tween the results of

'ECLTS' and those by the fini te el-rent m thed (us ig the 'A';3TS '

code). Additicnalc., the 'ECL 3' e thod consis ten t'.y calc;.11:es a g eater ba'* "u than the fini te e!-rent ~e thod, that is 'SJLTS'

s ccnse. rative.

"ECLIS" as described above has been i~plerent c-i for de tennining the bolt desi e m loads for routine applicat:crs. c.e progr':n regn res c l a t e dinens icns , ntr er o f bel t s, bol t size, bolt sca cing . 531 t stiffness, the app 1ied forces and the al!cuable bolt shear and t ens i on lo ads as irput s. Tne allowable ! cads for a given bolt are deternined based on the concrete edge dista .ce, bolt spacing, cm-403 04

1-079-1 2-079-1 h!r. r!. V. Sey f ri t July 6, 1979 beient length, shear cone overlapping, nnnufacturer's uit innte capacity, and a design safety factor (see paragraph o. 2 f or dis-cussion of alloc.nble loads) . 'Ihe prograni ccrputes the bolt forces and calculates a shear-tension interaction value based on the allcwable loads.

~he shear-tensicn interaction in the ancher holts has been accounted for. 'Ihe total applied shear is regaired to be carried by the bolts in accordance wi th the folIcwing interaction focula.

('IC/TA)' + (SC/SA) i 1.0 hhere IC = Calcula ted tens ile force TA = Allcwnble tensile force SC = Calculated Shear force SA = Allowable Shear force Thi s f a nnula i s reccnrnanded for bol t ed j o in t s b'f the 19'. 7 edi t ion o f AS.E Code ,Section III, Appendix r/II-2461.3. 'Ihis is one of the interacticn optiens wi thin the " DOLTS progran and has been used on the anjarity of our calculations. Sarce of our earlier calculaticns utilized a r: ore conservative interaction option of using the 5/3 power vice the 2 power.

(ii) For special cases where the design of the support didn' t lend itself to the foregoing method, the finite elenunt cuthod using the

'A';SYS" ccde ancilor other standard engineering analyt ical tech-niques wi th conservat ive asmu:ptions were eployed in the analysis.

(iii) C:her cases were solved using an approach based on the s t reng th de s i gn ac thcd g i . en in ACI 313 -77 ccde .

( i-) T.7, inch d: re ter piping sys:cs (and less) were analyzed wi th a cha: t cc:hcd . ice a cmputer prcgrr. ..e did inc'.2de these in cur inspecticn progran, see paragraph 4, and are in the process of analyzing the 'tpical base plates" authorized by the chart

- e : hod . We anticipate having the analysis ceple ted by 1 Augus t 1979.

"he current s tatus of this (re) analysis ef for: is tabulated below.

LNIT 1 C;IT 2 Total N3. of affected supports 493 537 To tal of supports (re) analyzed 372 359 To tal supports wi th unacceptable boIt loads 36 33 408 048

. v 1-079-1 2-079-1

!.tr. K. V. Seyfrit Ju!'f 6, 1979 The following break down is provided for the " supports with un-acceotable bolt lo ads . "

U'iIT 1 C'iIT 2 G I.SLNED

• ACI'ICN JUSTIFICATIGi SAFEIY FACICR

%F.LCE SHELL 29 14 4> SF> 3 5> SE> 3 None See Paragraph 2 (meets pro-posal ACI 349-76 Augus t 1973) 31 16 3 > SF> 2 3 > SF > 2 Will Ccnduct an inspect ion See paragraph 2 (see paragraph 4 (Meets A?&L for Dx Da tes) Position) 26 3 2 > SF 2 > SF Pedesign See Paragraph Support 4 for due dates t'e antic:pate car.ple ting our analytical ef fort by Augus t 1, 1979. We will subni t the results of the remining analyses by August 15, 1979. It should be noted that our schedule for analytical work on base plate flexibility extends beyond the Eulletin reporting time fr re of 6 July 1979. Tnerefore, w have already started our anchor bolt verification program, as described in paragraph 4.

T7 m.i Verify that the ccncre te expansicn anchor bolts have the following rni.,i un f ac tor o f sa fe ty be twen the bol t des ign load the bolt uti~nte capacit, dete =ined fr r static load tests.

(a) Fou- - For wdge r.d sleeve type anchor ' col t s ,

(b) Fi ve - Fo r shel 1 type a-.chor to'. t s.

Pl.SFC M E Tne bol c allowables utilized in these analyses are based on testing conducted at the Fist Flux Test Facili ty and on the re-spect ive ccncrete expe.n.sian ar.chcr mnuf acturer's suppli ed data.

These allcw tble forces account for eiedmnt length, r.inim: bol t s.nacin~5 and free edge distance.

• See At tachrmnt 2 for descriptic, of mthed for calcu'.ating safety factors.

408 049

s .c .

1-079-1 2-079-1

.'.ir . K. V. Sey f r i t July 6, 1979 In the current design review, factors of safety (i.e. ratio of bolt ultimate capacity to design lond), four for wedge type and five for shell type anchor bolts were used for service load cases. nhen extree envi rcrr. ental loads are included, a factor of sa'ety of three is acceptable in accordr.ce wi th Section B.7.2 of the Proposed Addi tien to Ccde Recuire ents for Nuclear Safety Ralated Concrete Structures (ACI 349-76) August 1973.

Further, where a specific support has been verified, a f actor of safety of txo is cons:dered to be satisf actorf with extrem envi rcnnent al lona present.

ITai Cesc ribe the design recui re ents, if applicable, for 2nchor bolts to wi thstand cyclic loads (e.g., seisrr.ic loads and high cycle operating loads).

RESKt;5E In the original design of the piping syst .a (Sechtel) c.. a-sidered deadw ight, then-al s tresses, seisrc.ic loa 6, anc ifna:nic loads in the generation of the pipe support design loads. To the extent that these loads include cyclic considerations, t: ese ef fects would be included in the design of the hangers, base pla .es and anchorages.

The safety f actors used for concrete expr.s:en anchors, ins talled cn succerts for safety related piping systes, wre not ir. crease for l o ads wh i ch are cyc ' i c in na t u re . Tne us e o f the s re s u e t" facter for cyc l i c r.d s t a t ic loads is based en the Fast Flux Tes: Facility Tests." Tne test r esul t s indi ca t e :

1. The e g r.s i:n ancho rs su cces s f a i l; wi ths t ecd r1 ~_i l l i ca rf c i e s of Icn; te m fatigue loading at a .u:i:nn ir te- si ty ci 0.20 cf the static ultimite capacity. then thercaxim lo a.a i r.t es i t e

• Cr i i l ed - In Expr.s ion Bol t s u .de r St at i c and Al t er .at n.; Ioads ,

Papo r t ?;o . ER- 5 3 5 3 4 b f Bech t e! Pc ca r Ccrp. , J r.uarf : 97 5.

408 0 ~> C

1-0 79 -1 2-079-1

?.'r . K . V. Sey f r i t July 6,1979 was steadily increased beyond the aforem2ntioned value and cycled for 2,000 times at each lo ad s t ep , the observed f ailure load was about the sant as the static ultimate capacity.

2. 'Ihe dynanic load capaci ty of the expansion anchors, under sinulated seistnic Icading, was about the sane as their cor-respondi on 7 static ulti:nate capacitier.

Im.iu Verify fran existing C~ doc rmatation that design require.ents have been nrt for each anchor bolt in the following areas:

(a) Cyc1ie loads have been consid 3 red (e.g. , anchor 501t pre-load is equal to or greater than bolt design load). In the case of the shell type, assure that it is not in contact with the back of the support plate prior to preload testing.

(b) Specified design size and type is correctly installed (e.g. , proper eiechunt depth) .

I f su f f i ci e it de curen t a t i on does no t exi s t , then initiate a testing program that will assure that rninintra design requirerrents have been tret wi th respect to sub-i tens (a) and (b) above.

RESFCME It is not necessa:f that the bolt preloaci be equal to or g eater than the bolt design Icad. Pipe stpports r.c. cnchors are subj e c t ed to s t a t ic anci br.ani: loacs. Tne d :ric lo W are sei r.ic loads which are short nu at:cn cy::ic loads. Tai s t.;e of cyclic

! cad :s not a f atigue loac, so the nrunt 0: preloc.d cn the bolts wi!I not g eati affect the perfo mance of the anchor:.ge. (In addi t ic n, cr . .e'. c ad is los t over the .

lite of the plant due to creep and ot,ner s:cu lar pnenccena) .

2neretore, : t.

t,ne :n t t : a,t ins *al-lation 'o =,ue on the bolt accaq)l: Shes the pu.gose of se' tin'a the wdge, then the ultinnte capaci ty o f the bol t is not affected by +he arrunt of preload present in the bolt at the t ire o f cyc '. i c lo adi ng.

For .ibrato:f Icads during plan

• operation, the e:gansion anchors have successfully wi ths tcod lone tenn fatiaue en'. i rcamnt as dis-0 cus s &i in the .crevious section.

m 40b 0b;

1-079-1 2-079-2

?.'r . K. V. Sey f r i t July 6, 1979 AlI concrete c:cansion anchors are designed, instal 1ed and vc_ t fied as per Speci fication 6600-2-C-2305 (At tached) for Uni t 2 and for Unit I additiens after June 23, 1974. Prior to this date, concrete expansion anchors for Uni t I were ins talled in accordance wi th rmnuf acturer's instructions. Installation, verification and testing procedu es along wi th acceptance criteria are given in Secticn 5.0 of this Specification.

Unit 1 A D-1 did not have suf ficient CC docinwntatien to veri fy that the design regairemats had been at for base plates installed prior to 23 June 1974. Tnerefore, we have initiated a testing progran which will assure that the r-ini: urn design regairemnts have been mt.

Selected CE3's were tested in accordance vd th Speci fication 11406-276-5 (Attached). Tne procedures described in Specification 11406-276-5 requires expansion anchors to be veri fied for loca-tion, elevation, nuier of anchor bolts, spacing and edge distance as shcv.n on design dreuings, type of anchor used, unbedmnt length and projection of anchors, washers, darnage to concrete, anchor bolt di ane t er and anchor bol t length. Also, expansion anchors are tested for Design Ioads using a srpling tecMigae specified in Secticn 3.0 of the speci fication. Since wc are not taking credi t for bol t preload, we are not presenting a correlacion between torg;e and teision. Chr test progran does shou that the ins tal lation torque on the bol t has accnplished the purpose of se t ting the wedge which de ter-

nincs the abili ty of the bol t to develop its ulti:nate capacity. Tne procer docum ntation, indicating the location of exnansion anchor and rcun repr esen t ed , a thod o f t es t (torc.ue or tension), test results, type of failu-e when applicable, date of test along with nane and signature of the inspector, : s ava il ab le a t the ;cbsi te.

A rr. dan sam;in; precedu e as per Specificati= 1;406-276-5 (At-tached) paragr ap h 4. 0 w a s us ed t o de t e n-i .2d

• i ch C2'd ' s to inspect.

H.e acceptr.ce criteria of 114M ~176-3 is a 75't con::dence level that there are less than 5% defectives in the total p:pulation. It should be noted that cur smpling technicue was developed r.d exe-cuted prior to the receipt of IE Sulle tin 79-02 Revision 1. 'Ihe r e ,

fore i t w,s a randan scrple en a pir.: basis . ice a syst e basis. This approach did ace t at least one base cla te f r= the n.i ori t., of the s /s t e-s and was in t ended to re f lec t the conditions of the pir.t.

'Ihus , we feel that it has a t the criteria of a representative sa pling technig:e. At tachmnt No. I pr ov ides a sunnry o f CE3 's t es ted pe r sys tcr.:.

c 40v

1-079-1 1-079-2

!.tr . K . V . Seyfrit -S- July 6, 1979 The results of our in-field inspection revealed that onlf I C8 out of the 122 tested f ailed the torque test portion of Specification 11406-276-5. This gives the "953 confidence level that our C 8's are correctly installed. Feuever, there we re 52 suppo r t s wh i ch we re found to have deviations fran the "as-tuilt' drawi ngs . These de-vi stions were of the following nature:

28 -

Discrepancies in anchor bolt size or type (24), base-plate dimensions (2) or structural arrang rent of the support (2).

24 -

Discrepancies in hanger location. These are jt.dged to have minimal effect on hanger loads, therefore, we do not p opose to rerun the hanger guidance at thi s t i:re .

V,'e have aircadf incorporated these discrepancies into our analytical work and inteid to continue our field ef forts until v.e achieve the 4% confidence level that our "as-built' pipe support drawings are cor rect . This extended ef fort wou'd be a 1003 verification of the Unit I seismic category I hanger sketches which emloy flexible baseplates and O 's. The following itens wi11 be verified:

a. Fanger location

b. Fanger structural arrangenent

c. Baseplate di:ensicns, including thichess

d. Eoli sice

e. Eolting patte m

f. Concrete edge distance, i f near the minimum for the bo'.t sice in cuestion.

Addi t i onal ly , we M l l be ext e-di ng the ori J na. 'in-f iel d' inspection pr m an to inc;ude the 31 Unit I r.d 1: En . t 2 CE's chich have

> 5.~ > 2, as repor ted in paragra h 1.

We anticipate that buth of these ef forts will be ca-ple ted duri .g the next re fueling cuta.;e and that any signi ficr.: discrepancie will be ccrrected at the firs t cppar* uni tf subsequent to the cutrge.

Uni : 2 A"O ha= suf ficient CC docxrntation 'o veri fy that the design regir n nts nace been art for each base plate. A revicu of the

'Speci tication for Ins tallatica of Class I and I'on-Class 1 Ccncrete Expa. m .:-Ivpc Anchors" (6600-2-C-2305) (At tached) and o f 'Fi eld Ins trum icn for Ins talla tion of Class ! Oncrete, E<pansien

-t e,

u bJ, U

1-079-1 2-079-1

?.!r. K. V. Seyfrit July 6,1979 Anchors (2FI-129) ( At t achnun t ) has u.;bs t an t i a t ed th i s f a c t .

The 'as-buil t " conforance proble.s discovered on AfD-1 are not present on eliO-2 because- 1) of these tm procedures, 2) a nuch s trcn;;er crphasis on CC vd th systen walk dcuns conduct ed by A-E and A%L personnel and 3) A'D-2 ut i1ised "on-sitc' en;;ineers to do the hange r design vi ce us ing a sub-cont ractor's "of f-s i te fa-cilities. For these reasons an extensive in-field pregran will not be rescired on AfD-2.

Ve ry t ruly yours ,

/' rD s' gy .- If , ,- (/ / ~aIb-~

[s David ManageC. r, LiTrinble ce .s ing DCr/3.GV/ ew cc:  ?.'r. .'!o nnal L. Moseley Director, Division of Reactor Operations Inspection

?uclear Regulatory Cmrission 4 3 50 Eas t Wes t Highway Eethesda, Maryland 20555 408 054

. . i ATTACH}ENT I UNIT 1 CONCE*E FXPANS.T'. ':0LT TESTING Not Seismic Cat. I Safety Syste: # Tested Related RCS Reactor Coalant 23 MU Make Up and Purificati:n 20 CF Core Flacding i DH Secay Heat Renoval 25 SMP Sampling 3 X FPC Spent Fuel Cooling 0 X MS Main Stean 3 ED Energency Diesel Cen. and Fuel Oil 0 FW Feedwater 22 R3S Reactor 31dg. Spray 2 AS Instrument and Servi:a Air 0 X ICC Intermediate Cooling 0 X SW Service Water 12 LW Caseous Radioactive Waste 1 CH Chilled Water 0 X FH Plant Heating 0 X CA Chenical Addition 0 X FW Fire Water 0 X Contain=ent Test Cenn. No Support H&V Heating and Ventilating Spare Centainment Flued Head No Support HP Heat and 7ent. (Hydrogen Purge Air Systen) 5

,s.

_ a ,s 3,,

a r--

413 8 055

ATTAC4'El.T 2 C2.31 3 SAFETt' FACIRS (GF)

In ecntmnicating the result of the CEE-base plate revie.v. it is nure straight forward to use GF than L-teraction Value (IV). IV is the result of the int raction equation. The GF tmy be cctputed based oi, the following for:ulas:

GF = M for wedge type CEB 5

GF = IV, for Ehillips Self-D-illing

&.chors and other shel1 t,7e Co.

403 Ob6

_